Produsing Theory in A Digital World 3.0 The Intersection of Audiences and Production in Contemporary Theory - Volume 3 (Rebecca Ann Lind (Editor) )

REBECCA ANN LIND, editor
produsing
theory in a
digital world
3.0
The Intersection of Audiences and

Production in Contemporary Theory
VOLUME 3
Continuing the explorations begun in the first two Produsing Theory
volumes, this book investigates some of the tensions generated in the
spaces enabled by the confluence of the formerly disparate activities
of producing and consuming media. Multiple and varied theories—
some still emerging—are invoked in attempts to illuminate the spaces
between what previously had been neatly-separated components
of media systems. This book is useful in a number of courses such as
media culture and theory, introduction to new media, the Internet and
the audience, new media theory and research, mass communication
theory, emerging media, critical analysis and new media, concepts of
new media, new media participants, new media in a democratic soci-
ety, critical studies in new media, new media and social media, dig-
ital media studies, participatory media, media audiences in a digital
world, digital cultures and social media, Web culture and new media
studies, introduction to new media, new media and society, and more.
“A collection of sparkling ideas from many of the field’s best thinkers,

this book is sure to generate productive discussion.”
—Nancy Baym, Senior Principal Researcher, Microsoft Research
“Through a series of thought-provoking essays bound to appeal to

those both in and new to the fields of media and communication stud-
ies, this volume minds the gaps between production, audience, and
Internet studies, bringing theories of the technological, computational,
discursive and critical along for the journey.”
—Lynn Schofield Clark, University of Denver
Rebecca Ann Lind (Ph.D., University of Minnesota) is Associate Profes-

sor of Communication at the University of Illinois at Chicago. She has
published on race, gender, class and media; journalism; new media;
media ethics; and media audiences.
w w w. p e t e r l a n g . c o m
119
REBECCA ANN LIND, edito
produsing theory in a digital world 3.0
produsing
Produsing Theory
theory
in a Digital World 3.0 in a
digital world
3.0
VOLUME
3
LIND, ED.
The Intersection of Audiences and

PETER LANG
Production in Contemporary Theor
VOLUME 3
Steve Jones
General Editor
Vol. 119
The Digital Formations series is part of the Peter Lang

Media and Communication list.
Every volume is peer reviewed and meets
the highest quality standards for content and production.
PETER LANG
New York • Bern • Berlin
Brussels • Vienna • Oxford • Warsaw
Produsing Theory
in a Digital World 3.0
The Intersection of Audiences

and Production in
Contemporary Theory
Volume 3
Edited by Rebecca Ann Lind
PETER LANG
New York • Bern • Berlin
Brussels • Vienna • Oxford • Warsaw
The Library of Congress has catalogued Volume I as follows:
Produsing theory in a digital world: the intersection of audiences

and production in contemporary theory / edited by Rebecca Ann Lind.
p. cm.—(Digital formations ISSN 1526-3169; v. 80)
Includes bibliographical references and index.
1. Mass media—Technological innovations. 2. Mass media and technology.

3. Mass media—Social aspects. 4. Media literacy. 5. Digital media. 6. Social media.
I. Lind, Rebecca Ann. II. Title: Producing theory in a digital world.
P96.T42P76 302.23—dc23 2012014097
ISBN 978-1-4331-1520-2 (Volume 1 hardcover)

ISBN 978-1-4331-1519-6 (Volume 1 paperback)
ISBN 978-1-4539-0840-2 (Volume 1 e-book)


ISBN 978-1-4331-5342-6 (Volume 3 epub)
ISBN 978-1-4331-5343-3 (Volume 3 mobi)
DOI 10.3726/b13192

Bibliographic information published by Die Deutsche Nationalbibliothek.
Die Deutsche Nationalbibliothek lists this publication in the “Deutsche
Nationalbibliografie”; detailed bibliographic data are available
on the Internet at http://dnb.d-nb.de/.

Cover image: Windsor by ChiTownMuggle (2019).
Reprinted with kind permission of the artist. All rights reserved.

© 2020 Peter Lang Publishing, Inc., New York

29 Broadway, 18th floor, New York, NY 10006
www.peterlang.com

All rights reserved.
Reprint or reproduction, even partially, in all forms such as microfilm,
xerography, microfiche, microcard, and offset strictly prohibited.
To my parents and my students.
Contents
Acknowledgments ix
1. Produsing Theory in a Digital World: Minding the Gap 1
Rebecca Ann Lind
2. “With and between you all”: Celebrity Status, User-Audience Networks,
and Representative Claims in Emma Watson’s Feminist Politics 11
Ellen Watts and Andrew Chadwick
3. L8r H8r: Commoditized Privacy, Influencer Wars, and Productive
Disorder in the Influencer Industry 31
Crystal Abidin
4. Production and Performance of White Anti-Racism in Online Media 49
Michael Potts
5. Networked Gatekeeping and Networked Framing of #BlackLivesMatter
Publics during the 2016 US Presidential Election 67
Sharon Meraz
6. The Potential of Social Media Groups to Afford Users a Voice 87
James Ngetha Gachau
7. “Glamorous factories of unpredictable freedom”: Care, Coalition, and
Hacking Hacking 105
Christina Dunbar-Hester
8. Religious Influencers: Faith in the World of Marketing 121
Mara Einstein
viii Contents
9. Audiences, Affects, Attachments: Theorizing Textual Approaches to

Digital Culture 139
Akane Kanai
10. Discourse-Analytical Studies on Social Media Platforms:
A Data-Driven Mixed-Methods Approach 157
Ehsan Dehghan, Axel Bruns, Peta Mitchell and
Brenda Moon
11. The Hyperperception Model: How Observing Others on Social
Media Can Affect People in Close Relationships 177
Erin L. Spottswood and Christopher J. Carpenter
12. Proposing a Model of Social Media Use and Well-Being 195
Marina Krcmar, Drew P. Cingel, Yifan Zhao, and
Lauren Taylor
13. Audiences Assemble: Becoming an Audience and Produser in Mixed
Media Environments 211
Annette Hill
14. Coordination, Continuity, Configuration: Toward a Mattering
Framework for Human-Machine Produsing 229
Jaime Banks
15. Afterword: The Legitimacy of Produsage 245
Robert W. Gehl
Contributors 257
Index 261
Acknowledgments
With the publication of this book, Produsing Theory has become a trilogy. Just
as in the first two volumes, my greatest thanks must be given to the contrib-
utors. I have enjoyed working with and learning from the contributors in this
and the first two volumes. Besides generously sharing their intriguing ideas
in the following pages, the authors have been responsive, understanding, and
still willing to look at my emails even after some fairly intense conversations
during the writing process. I hope they are pleased with the outcome.
In addition, and as always, many thanks are due to my colleagues (espe-
cially Steve Jones and Zizi Papacharissi) in the Department of Communication
at the University of Illinois at Chicago, and to the College of Liberal Arts and
Sciences (and its Dean, Astrida Orle Tantillo), for their support.
The folks at Peter Lang have also been a pleasure to work with, and I owe
special thanks to the three editors who have been involved in the Produsing
Theory books: Erika Hendrix, Kathryn Harrison, and Mary Savigar. Thank
you for helping bring these ideas to fruition!
Grateful acknowledgment is made to the following for permission to use
copyrighted material:
Cover image: Windsor by ChiTownMuggle (2019). Reprinted with kind
permission of the artist. All rights reserved.
1.
Produsing Theory in a Digital
World: Minding the Gap
Rebecca Ann Lind
Minding the gap. This book, like the other two Produsing Theory volumes, is
all about the gaps. But unlike the physical and perilous gaps of public trans-
portation systems such as the London Underground, the gaps with which we
are concerned are conceptual and productive—they’re the spaces between
what previously had been neatly separated and discrete components of media
systems. In some settings, division by audience, content, and production set-
tings remains useful (e.g., Lind, 2017), but this volume, like the previous two
(Lind, 2012, 2015), focuses on the interstitial spaces and the intersections
between and around these settings: the gaps.
We may still refer to texts, audiences, and producers, but in a media envi-
ronment increasingly driven by what Coates (2005) called social software, we
must expand our focus if we are to understand the relationship among these
components. Coates defined social software loosely as “software which sup-
ports, extends, or derives added value from, human social behaviour—mes-
sage-boards, musical taste-sharing, photo-sharing, instant messaging, mailing
lists, social networking.”
In this book, we’ll continue our exploration of the gaps such as the spaces
and intersections between self and other (including nonhuman others),
between audiences and texts (including how audiences assemble), between
exploration and inhibition (including reinforcing or challenging hegemonic
racial or gendered identities), and—of course—between the production and
use of media. Each of these gaps is a site that can be illuminated by the various
theories and methods, some still emerging, presented herein. The chapters
present multiple perspectives to consider and study these spaces made pos-
sible by ongoing developments in communications technologies and social
software. Each has some type of connection to what Axel Bruns (2008) called
2 Rebecca Ann Lind
produsage, briefly defined as “the collaborative and continuous building and

extending of existing content in pursuit of further improvement” (p. 21).
Produsers enact a hybrid role in the system—they both produce and use
media content. They thrive in the gaps.
There are many ways to organize this book’s chapters. The table of con-
tents presents one way. But, as with the prior volumes, in what follows I shall
follow Papacharissi’s (2012) advice to play with linearity, rather than proceed
in a numerically sequential fashion through the chapters. So let’s see what
is revealed by approaching the chapters based on the spaces they illuminate.
The chapters reveal many gaps and intersections other than these, and share
significantly more overlaps than space allows me to consider here (I encour-
age you to review the introductory chapters of the first two volumes). For
now, we’ll first consider the chapters that address gaps in methods. Second,
we’ll look at the chapters having a primary goal of addressing gaps existing
in and around theory by presenting preliminary theoretical models. Finally,
we’ll turn to the chapters addressing a different sort of gap in theory and
the extant literature by focusing on specific groups (including celebrities and
other socially constructed groups such as those based on race or gender).
All chapters in this book, as do those in the other two Produsing Theory
volumes, contribute to our understanding of the communication processes
facilitated by social software. The contributors were chosen to represent a
variety of paradigms and perspectives. The chapters present diverse insights as
afforded by the humanities, critical/cultural studies, and the social sciences.
When combined, they facilitate understandings not readily encouraged by
reliance upon a single vantage point.
As noted, many chapters specifically strive toward developing theory and
method. For example, Ehsan Dehghan, Axel Bruns, Peta Mitchell, and Brenda
Moon (Chapter 10) present a solid argument for the need to mind the gap
between qualitative and quantitative methods, which have dichotomized, for
example, discourse analytic studies and studies using so-called big data or big
social data. The authors present a cyclical mixed-methods approach of social
media discourse analysis, focusing on two case studies rooted in political dis-
cussions in the Australian Twittersphere: #RoboDebt, what the authors call
“an algorithmic scandal” related to automated attempts to recover poten-
tial overpayments made to welfare recipients, and a long-standing debate
over whether a section of Australia’s Racial Discrimination Act should be
repealed or revised. These two cases demonstrated how Dehghan et al’s inter-
sectional methodology can reveal rich insights into multiple aspects of the
communication content and context. Besides the value of the actual results
of these case studies—which revealed the presence of competing antagonistic
Produsing Theory in a Digital World: Minding the Gap 3
discourses—the authors show how their cyclical mixed-methods approach can

help discourse analysts select meaningful samples from large-scale data sets.
Dehghan et al’s chapter complements the methodologically oriented
contributions found in Volume 2 of Produsing Theory. For example, Darryl
Woodford, Ben Goldsmith, and Axel Bruns (Chapter 9) presented social
media metrics as a new form of audience measurement. Heidi Vandebosch,
Philippe Adam, Kath Albury, Sara Bastiaensens, John de Wit, Stephanie
Hemelryk Donald, Kathleen Van Royen, and Anne Vermeulen (Chapter 14)
argued for the need to engage adolescents in narrative research and interven-
tions on cyberbullying.
Multiple chapters in the present volume present preliminary theoretical
models. Focusing on the realm of interpersonal communication supported
by social software, Erin Spottswood and Christopher Carpenter offer what
they call the hyperperception model. This model theorizes how and why peo-
ple might read too much into (hyperpercieve) the social media interactions
between their significant others and unknown others. The authors present the
results of a preliminary test of the model, in which nearly one-third of partic-
ipants revealed that their romantic partners had indeed over-interpreted and
inflated the meaning of at least one social media post with an unknown other.
Participants reported that essentially benign posts and comments on a social
media site had caused discord both within (when the partner jumped to con-
clusions about or became jealous of the unknown other) and beyond (when
the participant felt the need to unfriend the unknown other) the romantic
relationship.
Bringing a different perspective to theorizing social media effects, Marina
Krcmar, Drew Cingel, Yifan Zhao, and Lauren Taylor (Chapter 12) incor-
porate variables related to social media use, processing, and effects into their
model of social media use and well-being. Krcmar et al’s model begins with
addressing individual differences, users’ motivations for using social media,
and platform affordances (e.g., image—or text-based social media). The
model also includes active and passive social media behaviors and cognitive
processing strategies, presenting the likelihood that each combination might
yield different outcomes. Social proximity between users is seen as a moderat-
ing variable, and various indicators of well-being can be considered outcomes.
Extending the very boundaries of produsing as it has traditionally been
considered, Jaime Banks (Chapter 14) presents a model for considering
human-machine produsing. Banks argues that a new class of actor—partic-
ular types of machine or assemblage, such as certain algorithms, networks,
devices, semi-autonomous agents, artificial intelligence, and more—must be
considered alongside human actors in the spaces between creation and use of
media content. The machines or assemblages with which Banks is concerned

have both agency and sociality; they create content and meaning in partner-
ship and in conjunction with human actors. Her Mattering framework is a
preliminary model for understanding human-machine associations at work in
generating and using media. It posits three dynamics of produsing that reflect
joint agency: coordination, continuity, and configuration.
Although the above-mentioned chapters explicitly present new models
and methods, all of the contributions function to extend theorizing about
produsing. I’ll mention just a few more here. Annette Hill (Chapter 13)
prompts us to consider what happens before produsing, that is, before people
mobilize to engage in produsage. Using theories of audiences, affect, and
assemblage, Hill situates her analysis in the dystopian TV drama Utopia. Her
study is richly rooted in the text itself, interviews with people involved in
creating the program, and interviews with audience members and fans of
the program. Paying careful attention to the juxtapositions of the show’s
soundtrack and visual storytelling, she interrogates how audiences assemble,
become aware of themselves as audiences, and, in essence, constitute them-
selves. As and after they assemble, these audiences can become what she calls
path makers as opposed to path followers, become alert to power relations,
be inspired to political and cultural engagement, and participate in remix
projects or other forms of produsage.
Akane Kanai (Chapter 9) urges us not to neglect textual approaches,
arguing their value in a digital culture that increasingly textualizes our iden-
tities and encourages us (especially young women) to perform and display
ourselves in particular ways, often invoking celebritized logics. Informed by
Livingstone’s text-reader metaphor and Berlant’s feminist theorizing of inti-
macy as regulatory and related to the dominant ideology, she presents how
audiences/produsers engage the discourses through which they are already
addressed. Kanai illustrates her argument with examples of how audiences
remixed and remade Jennifer Lawrence and Amy Schumer to reinforce nor-
mative femininity as authentic and natural, while masking the labor required
to perform narrowly defined ideals of femininity.
The varied theoretical contributions of these chapters combine to illu-
minate many gaps in the produsing space. They are welcome additions to
the theorizing evident in the first two Produsing Theory volumes. Volume
1 included Jay Bolter’s work on procedure and performance (Chapter 3),
Shayla Thiel-Stern’s piece on youth identity and performance and breaking
the fourth wall online (Chapter 6), Brittany Lee and Lynne Webb’s Identity,
Content, and Community (ICC) model of blog participation (Chapter 11),
and Catherine Heilferty’s consideration of online communication during
childhood cancer (Chapter 10). That volume also included Matt Hill’s theo-
rizing of spoilers and fans’ self-narratives (Chapter 7), Diego Costa’s analysis
of the cuckold fantasy in amateur porn as colonial encounter (Chapter 9), Eric
Freedman’s work on technobiography and the networked body (Chapter 4),
Paul Booth’s work on demediation and playful ideology in the role-playing
adventure game MagiQuest (Chapter 5), and Axel Bruns and Tim Highfield’s
contribution on the produsage of citizen journalism (Chapter 2). In Volume
2, we find Dmitry Epstein’s work on the structuration of internet governance
(Chapter 3) and Jeremy Hunsinger’s theorizing on darknet consummativi-
ties (Chapter 4), John Pavlik’s use of flow theory (Chapter 6), and Thomas
Lindlof’s (Chapter 2) consideration of interpretive communities in the Web
2.0 environment. Nicholas Bowman theorized the cognitive demands asso-
ciated with video gameplay (Chapter 7), Bradley Gorham and Jaime Riccio
presented a preliminary model for social media’s impact on adolescent iden-
tity formation (Chapter 5), and Philip Napoli and Jonathan Obar (Chapter 8)
argued that the mobile conversion is leading to what they call a repassification
of the audience. Kishonna Gray explored the liberatory potential of Black
cyberfeminist theory (Chapter 11), and Annette Markham looked at produs-
ing ethics, or the everyday production and negotiation of ethics (Chapter 15).
Another way to approach the chapters in the present volume is to con-
sider their recurring themes, one of which is celebrity (as already noted in the
discussion of Kanai’s Chapter 2 contribution). Other chapters also connect
to traditional and microcelebrities or internet influencers. For example, Ellen
Watts and Andrew Chadwick (Chapter 2) analyze how Emma Watson nego-
tiated her engagements with her feminist book group and discussion forum
Our Shared Shelf. Informed by Saward’s theory of representative claims and
Bourdieu’s theories of fields and capital, Watts and Chadwick shed light
on the gaps between status differences (celebrity vs. co-participant), enter-
tainment and politics, and proximity and distance. The authors argue that
Watson was able to exercise political power and represent the group politi-
cally because she successfully performed multiple claims to represent user-au-
dience networks. By emphasizing her role as a facilitator, she was accepted
as a connected representative; by positioning herself as a fellow learner, she
was accepted as an ordinary member of the group. Finally, by using social
media to connect with the group—posts that were shared and picked up by
online news outlets because of her celebrity capital—she became accepted as
an authentic ambassador.
Crystal Abidin (Chapter 3) considers so-called Influencer wars (conflicts
among or between professional microcelebrities who thrive on the internet).
Drawing upon multiple theoretical perspectives, she argues that Influencer
wars can be understood as social dramas, pseudo-events, a form of productive

disorder, and negative attention events transpiring in an attention economy.
Abidin presents case studies of three genres of Influencer wars (status claims,
authenticating appearance, and tell-all exposes). In the process, she eluci-
dates the concept of commoditized privacy and presents what she calls “web
amnesia,” which acknowledges a tension between the persistence of material
posted to a social network (boyd, 2011) and the perceived ephemerality of
the staged spectacles comprising Influencer wars.
Mara Einstein (Chapter 8), who argued that religion and marketing have
much in common, focuses on religious influencers. Noting that social media
practitioners can have influence beyond the capital market, and aware of the
importance and personal nature of religion, she presents a preliminary typol-
ogy of religious influencers. The typology reflects a broad range from those
who are purely focused on faith (Traditional Religious Leaders) to those who
simply use faith as a pitch to sell products (False Profits). In a variant of the
citizen-consumer dichotomy—which has very real implications for audience
members in a democratic society—Einstein wonders whether the followers
of highly market-oriented influencers might begin to see themselves not as
congregants or seekers but as consumers.
A second recurring theme in this collection addresses issues of particular
socially constructed groups, such as those based on gender or race. Most of
these also connect in some way with activism. As is already evident, the chap-
ters by Watts and Chadwick (Chapter 2) and Kanai (Chapter 9) foreground
gender. So, too, does Chapter 7, by Christina Dunbar-Hester. Dunbar-Hester
engages with the open technology movement, especially the free/libre and
open source software community, which has traditionally been overwhelm-
ingly male. A number of these groups have acknowledged this limitation and
overtly attempted to reconstitute themselves in a way that is more welcoming
to diverse others. Reaching out to women might seem an obvious and appro-
priate strategy, but as Dunbar-Hester notes, reaching out only to women (and
not, for example, to women of color or genderqueer people) can inadver-
tently perpetuate both gender essentialism and the notion of gender as binary.
She argues that specific and limited calls (such as for women to participate)
both represent and constitute the communities, reflecting the I-methodology
of designers who consider themselves representative users. Dunbar-Hester
suggests some approaches for hacking communities that strive to overcome
the challenges of increasing their diversity.
James Gachau (Chapter 6) considers both gender and race in his study of
whether Facebook groups can challenge the dominant ideology even as they
exist within a hegemonic public sphere. Using Couldry’s concept of voice,
Gachau investigates whether the affordances of social software can allow users
to tell truly authentic stories and create voices for themselves as members
of certain distinct, and distinctly marginalized, groups. Focusing on three
well-organized Facebook groups (Freethinkers Initiative Kenya, Pan-African
Network, and Women Without Religion), Gachau found that the participants
could indeed use Facebook to create authentic narratives of their subaltern
lives, such as those about being oppressed by patriarchal religion or about
institutionalized discriminatory practices against Blacks and other people
of color.
In Chapter 4, Michael Potts presents White anti-racism as behavior with
complex ties to issues of presenting and policing White identity, controlling
the dialogue around racial inequality, claims of attempting to tackle racial
inequality, and conflating race identity and class identity. A number of online
media outlets manufacture outrage by presenting a racist remark or action out
of context, whipping up condemnation about the remark, and then report-
ing on the condemnation. For the media outlets, such emotion-stirring prac-
tices function to increase page views and shares in a click-driven advertising
economy. However, the act of sharing these stories is also functional for the
social media users: it displays their values—presenting the sharers as anti-rac-
ist—and it shows that the sharers care about the issue at hand. In the process
the racist behaviors often become attributed to other Whites—Whites unlike
themselves, frequently those considered poor White trash. More deeply, Potts
argues, this form of White anti-racism allows Whites to contain dissonance
and to rationalize their privilege. They may rail against the actions or lan-
guage of the individual Whites they have presented as other and as unlike
themselves, but in doing so they have not only failed to do anything about
the problem but also have reinforced the displacement of the focus from
themselves to the other.
Sharon Meraz (Chapter 5) applies theories of networked gatekeeping
and networked framing to investigate the digital activism associated with the
#BlackLivesMatter hashtag on Twitter in the year leading up to the 2016
US presidential election. For her analysis of networked gatekeeping, Meraz
identified the users who most frequently tweeted with the #BlackLivesMatter
hashtag (i.e., prominent users) and found that a relatively small number of
users were responsible for the majority of the tweets. She also discovered
that bots and suspicious accounts were prevalent on the #BlackLivesMatter
hashtag. In her analysis of how the hashtag enabled networked framing, Meraz
discovered multiple overarching frames characterizing the #BlackLivesMatter
movement, each of which was supported by or functioned alongside a set of
lower level hashtags. Although most of the frames supported the movement,
others attempted to delegitimate or even troll #BlackLivesMatter. By tracking

the use of the lower level hashtags month by month, and combining attention
to both the higher level and lower level frames, we can understand the net-
worked framing of #BlackLivesMatter as simultaneously constant and fluid.
The higher level frames focus on major issues, whereas the lower level frames
address specific concerns that may wax and wane over time.
These chapters, focusing on certain social groups, complement a num-
ber of chapters in the prior two Produsing Theory volumes. For example, in
Volume 1, Gust Yep, Miranda Olzman, and Allen Conkle (Chapter 8) studied
progress narratives, politics of affect, and queer world-making in the “It Gets
Better” project. Diego Costa (Chapter 8) highlighted race in his analysis of
amateur porn. In Volume 2, Kishonna Gray looked at race, gender, and virtual
inequality (Chapter 11). In Volume 2 chapters approaching social software as
a tool, Ella McPherson (Chapter 12) considered how civilian witnesses can
surmount the verification barrier in digital human rights reporting, Radhika
Gajjala, Dineh Tetteh, and Anca Birzescu looked at how the subaltern self
and the subaltern other are staged in information and communication tech-
nologies for development (Chapter 10), and Renee Hobbs used Twitter as
a pedagogical tool in higher education. Several Volume 2 chapters, already
noted, focused on adolescents: Vandebosch et al. (Chapter 14) and Gorham
and Riccio (Chapter 5); these are joined by Thiel-Stern’s chapter (Chapter 6)
in Volume 1.
Taken together, the chapters in this third volume of Produsing Theory
(coupled, of course, with those in the prior two volumes) help us produc-
tively mind the gap. Focusing multiple spotlights onto the intersection of
audiences and production made possible by social software facilitates a more
nuanced perspective than would otherwise be possible. However, what we
have learned here also reveals more gaps, and more questions.
I hope you enjoy the process of finding answers, and uncovering ques-
tions, as you read what follows.
References
boyd, d. (2011). Social network sites as networked publics: Affordances, dynamics, and
implications. In Z. Papacharissi (Ed), A networked self: Identity, community, and cul-
ture on social network sites (pp. 39–58). New York, NY: Routledge.
Bruns, A. (2008). Blogs, Wikipedia, second life, and beyond: From production to produsage.
New York: Peter Lang.
Coates, T. (2005, January 5). An addendum to a definition of social soft-

ware. Retrieved from http://www.plasticbag.org/archives/2005/01/
an_addendum_to_a_definition_of_social_software/
Lind, R. A. (Ed.) (2012). Produsing theory in a digital world: The intersection of audiences
and production in contemporary theory. Digital Formations series, Vol. 80. New York,
NY: Peter Lang.
Lind, R. A. (Ed.) (2015). Produsing theory in a digital world 2.0: The intersection of audi-
ences and production in contemporary theory. Volume 2. Digital Formations series,
Vol. 99. New York, NY: Peter Lang.
Lind, R. A. (Ed.) (2017). Race and gender in electronic media: Content, context, culture.
Electronic Media Series, Vol. 6. New York, NY: Routledge.
Papacharissi, Z. (2012). Afterword: A remediation of theory. In R. A. Lind (Ed.),
Produsing theory in a digital world: The intersection of audiences and production in
contemporary theory (pp. 189–197). New York, NY: Peter Lang.
2.
“With and between you all”: Celebrity
Status, User-Audience Networks,
and Representative Claims in Emma
Watson’s Feminist Politics
Ellen Watts and Andrew Chadwick
As part of my work with UN Women, I have started reading as many books

and essays about equality as I can get my hands on … I decided to start
a Feminist book club, as I want to share what I’m learning and hear your
thoughts too.
—Emma Watson (Our Shared Shelf, 2016)
In September 2014, Hollywood actor Emma Watson stood in front of the UN

General Assembly to invite citizens to “step forward” and “speak up” against
gender inequality by supporting the UN Women HeForShe campaign (UN
Women, 2014). Her speech attracted attention from news and entertainment
media around the world, and the HeForShe conference was watched online
more than 11 million times (HeForShe, 2015). The scale of the attention
across digital platforms led Twitter to paint the hashtag #HeForShe on a wall
at its headquarters (Nichols, 2014). Watson’s public persona had long been
intertwined with Hermione Granger, the studious and steadfast character in
the 2001–2011 Harry Potter film series—the second highest grossing enter-
tainment franchise of all time (Forbes, 2017). The social media followings
Watson had accumulated also meant she was well placed to deliver the reach
that the UN Women organization had hoped for (BBC Newsbeat, 2014). On
her Instagram account, for example, Watson’s work representing UN Women
and meeting world leaders sits alongside posts promoting her films and mod-
eling high fashion.
Closer scrutiny of Watson’s social media posts soon reveals that she has
gone beyond a conventional UN role in her efforts to promote feminist causes.
In January 2016, Watson launched Our Shared Shelf (OSS), a feminist book
group and discussion forum hosted on the Goodreads platform. On reaching
12 Watts & Chadwick
the milestone of 100,000 members within a month, Watson (2016d) thanked

the book group for members’ “heart warming” contributions and promised
to “keep going out there … to make this the best it can be.”
The connections provided by Watson’s celebrity capital enabled her to
contribute to the OSS book club in ways that most of its members would find
impossible. As we show in this chapter, Watson not only promoted the group
across digital platforms, but also attracted broader media coverage to it and
women’s rights more broadly. Watson’s celebrity capital was evident as she
secured interviews with feminist authors on behalf of OSS, providing a point
of connection between members and public figures. By January 2019, OSS
had grown to over 220,000 members and had hosted discussions on topics
ranging from feminist literature to personal experiences of sexual discrimi-
nation. Watson framed her decision to start the group in the context of her
formal UN role, telling prospective members she wanted to “share what I’m
learning” and “hear your thoughts too” (Our Shared Shelf, 2016).
These stated aims of interaction and sharing, however, potentially placed
Watson, the movie star and UN ambassador, in close proximity to those who
responded to her call to “join up and participate” (Our Shared Shelf, 2016).
This raises the question of how Watson’s celebrity status actually works in an
online community grounded in collaboration and community building, and
how she manages her relationship with audiences who sometimes become
co-participants, or what we term user-audience networks (Chadwick, 2017;
Chadwick, O’Loughlin, & Vaccari, 2017). Understanding how these pro-
cesses play out matters because the response of user-audience networks is
today central to how celebrities achieve the legitimacy, the authority, and ulti-
mately the power to switch back and forth between the fields of entertainment
and politics. We argue that the ability to translate the celebrity capital gener-
ated through entertainment media representations into the political capital
required for advocacy and mobilization for political ends is built on claims
to represent user-audience networks (Driessens, 2013; Saward, 2010). Our
approach to the relationship between celebrity and politics therefore places
celebrities’ modes of interaction with user-audience networks at the center
of explaining how celebrities migrate into the political field. To obtain the
political legitimacy required to advocate for feminist causes, Watson needed
to gain, and continuously maintain and renew, the acceptance of user-audi-
ence networks. Doing so, however, required that she avoid accusations that
she was inauthentically stage-managing this process from above, for her own
personal or reputational gain.
In this chapter, we blend interpretive and digital ethnographic methods
to show how Watson performed three types of claim to represent user-audi-
ence networks and, in turn, how these claims were evaluated by members of
Celebrity Status and User-Audience Networks 13
those networks. We show that Watson’s activity on the OSS forum allowed
her to act in close proximity to co-participants as an ordinary member of the
forum, while simultaneously creating the social distance that was required
for her to be the group’s connected representative. Watson was actually more
visible as the group’s external representative when she used her activities
beyond the group, particularly her social media posts, to assume the role of
authentic ambassador for the group’s feminist ideas. We argue that Watson’s
framing of OSS as a discussion “with and between you all” (Our Shared
Shelf, 2016) was a carefully formulated rhetorical move. This phrasing man-
aged the contradiction between, on one hand, Watson’s minimal levels of
direct engagement with others on the OSS group and, on the other hand,
her role as a representative of the group. Interviews with ordinary OSS
members show that it was precisely Watson’s negotiated distance from the
everyday entanglements of interaction with user-audience networks that
underpinned OSS members’ comfortable acceptance of her as a political
representative.
Although Watson’s celebrity capital supported her representative claims
by affording her considerable reach on social media, this capital alone could
not facilitate her acceptance as a legitimate representative. It was her connec-
tions with formal politics in the UN, together with the perceived appropri-
ateness of her professional self-presentation and engagement at a distance,
which enabled OSS members uncomfortable with celebrity to accept and
support Watson as a worthy exception. In contrast with the view that digital
media place celebrities and audiences in close proximity to each other, by
blurring the boundaries between media production and consumption (for
example Jenkins, 2006), we show that social distance and boundary mainte-
nance remain key resources that enable entertainment celebrities to act in the
political field.
Celebrity and Digital Media: Representing Proximate

User-Audiences
We cannot think about celebrity without thinking about audiences. As
Driessens (2013) argued, celebrity capital is a resource accumulated through
recurrent media representations and can be exchanged or translated as part
of a strategy to move between fields. Although celebrity is produced by and
through media representations (Rojek, 2001; Turner, 2014), in reality pro-
duction cannot be isolated from consumption because audiences—and the
celebrity’s representation of those audiences—are central to celebrity power.
Celebrities, Marshall convincingly argued (2014, p. 244), are required to
“somehow embody the sentiments of an audience.”
14 Watts & Chadwick
The relationship between celebrities and audiences has, of course, evolved

in significant ways over the last decade. Many audience members now pub-
licly produce, or co-produce, the symbolic resources upon which celebrity
depends. Whereas mass-media representation underpinned the growth of
contemporary celebrity culture (Rojek, 2001), some types of celebrity status
no longer require mass media or targeting a mass audience. Social media
platforms have enabled what are called “micro-celebrity” practices—where
followers are seen as an audience of fans regardless of how many are watching
(Marwick, 2013; Marwick & boyd, 2011)—and new forms of celebrity that
are native to the internet (Abidin, 2018). But although these developments
may have altered the balance of power between some celebrities and their audi-
ences, it is clear that interactions between celebrities and audiences remain the
key to understanding why celebrities can come to exercise political power.
Micro-celebrity practices have also found their way into how elite celebri-
ties behave. Everyday efforts to attract attention and build a following online,
such as responding directly to followers and sharing personal information to
give the impression of intimacy, have become essential parts of the repertoires
of many mainstream celebrities (Marwick & boyd, 2011). Emma Watson has
certainly been successful in cultivating a social media following. By 2019, she
was rapidly approaching 50 million followers on Instagram, placing her in
the top 35 on the platform worldwide. She also had 29 million followers on
Twitter and 34 million followers on Facebook.
Whereas most fans could previously only engage in parasocial interac-
tion with entertainment stars (Horton & Wohl, 1956), many now engage
directly. Turner has argued that the direct interaction enabled by social media
“inevitably reduces the distance” between celebrities and audiences (2014,
p. 75). And yet, celebrities such as Watson possess resources to manage fame
and retain some distance from their audiences. They can choose to make
fewer disclosures about their personal lives than internet celebrities who, as
Abidin (2018) has shown, hold greater obligation to the audiences they have
cultivated. Turner (2014) cautioned that although celebrity is more widely
distributed today, it is still based on hierarchical relationships. Not all celebri-
ties engage in constant contact with their fans; indeed the “honorific status”
of celebrity, as Rojek described it, is often based on elevation, social distance,
and a lack of “direct, personal reciprocity” (2001, p. 12). The key question,
as we see it, is how celebrities are able to retain the distance associated with
higher status while still representing user-audience networks on social media.
We suggest that this balancing act is particularly important for a celeb-
rity who wants to develop claims to represent people politically and exercise
political power by mobilizing in favor of a cause. Representation is a key part
of Bourdieu’s theory of how political actors compete for power. Competition
in any field is based on the volume and composition of capital that an agent
possesses. Types of capital vary in value, with a current type corresponding
to each field as a main power or stake (Bourdieu, 1987). Competition in the
political field is competition for the power of mobilization that is an essen-
tial part of political capital (Bourdieu, 1991). Whereas other forms of capital
such as economic, cultural, or social are exchangeable for movement within
or between fields, symbolic capital—the “recognition” obtained within a par-
ticular field—is the form capital takes when it is “perceived and recognized
as legitimate” (Bourdieu, 1987, p. 4). Symbolic capital in the political field is
not simply recognition, but the recognition an agent receives from a specific
group. Political capital is specifically derived “from the trust a group places”
in the politician. Recognition and credibility in the political field therefore
exist “only in and through representation, in and through trust, belief and
obedience” (Bourdieu, 1991, p. 192). Although the broader legitimacy of
celebrities is always connected to their audiences, we argue that celebrities’
ability to obtain political capital requires that they be perceived as representing
that audience in the political field.
We augment this insight from Bourdieu with Saward’s argument that
representation is not a so-called static fact confined to electoral politics, but is
performed through claims “to represent or know what represents the inter-
ests of someone or something” (Saward, 2010, p. 38). Saward argued that
representative claims are legitimated through acceptance by what he terms
“appropriate constituencies”: those who are invoked or who consider them-
selves to be implicated in a claim (p.148). This places audiences at the heart of
political recognition; indeed, representative claims cannot exist unless “audi-
ences acknowledge them in some way” (p. 48). Such acknowledgment, where
expressed as acceptance, empowers a celebrity to act politically.
To exchange celebrity capital for political capital, therefore, Watson
needed to construct claims to represent certain groups of citizens as she inter-
vened in the political field. Watson’s large social media followings certainly
lend support to such claims; yet too much engagement with audiences might
undermine the social distance associated with elite celebrity status. This bal-
ancing act between proximity and distance is further complicated in the con-
text of OSS, as we now discuss, due to the community-oriented affordances
of online message forums.
Community and Celebrity: The Affordances

of Our Shared Shelf
Standing before the UN General Assembly in September 2014, Emma
Watson was “reaching out” to the millions who watched her speech online
16 Watts & Chadwick
because, as she said, “we need your help” (UN Women, 2014). Watson situ-
ated OSS within her role as a UN Goodwill Ambassador, telling prospective
participants she wanted to “share what I am learning” from reading “as part
of my work with UN Women” (Our Shared Shelf, 2016). By 2019, 1.7 mil-
lion people had taken the UN’s “HeForShe commitment” by completing
a form on the campaign’s website, pledging to “take action against gender
bias, discrimination and violence.” The campaign claimed to have sparked 1.3
billion “social media conversations” (HeForShe, 2016), even though it did
not afford obvious opportunities for supporters to communicate with each
other. Although HeForShe’s website provided resources and ideas for those
seeking to “take action,” it lacked a dedicated platform to share ideas or infor-
mation. In practice, the structure, aims, and affordances of Watson’s online
feminist book group and discussion forum varied significantly from UN
Women’s HeForShe campaign. OSS afforded greater opportunity for citizens
to communicate, collaborate, and build networks. But this presented Watson
with tensions to negotiate as she performed claims to represent user-audience
networks.
Watson wanted to “share” what she was learning, yet she told prospective
members: I want to “hear your thoughts too” (Our Shared Shelf, 2016).
When the group reached 100,000, Watson (2016d) described her pride in
the burgeoning community she perceived, praising the “amazing … level
at which I see these topics being engaged with and discussed.” The forum
provided spaces for members to discuss the books selected on a bimonthly
basis—usually by Watson—and to contribute to discussions on a broad range
of topics related (and unrelated) to feminism. Other sub-forums provided
space for members to arrange meetups, pass books on to others, and suggest
ideas for the group or books for selection. Beyond OSS, the affordances of
the Goodreads platform encourage discussion and connection between mem-
bers, who can add each other as “friends,” leave comments on their own or
friends’ profiles, and send and receive private messages. OSS is publicly vis-
ible, but participation requires a Goodreads account and joining the group.
This, combined with the visible moderation of the forum, has specific
benefits for people seeking to engage with feminist discussion online. The
affordances of social media platforms such as Twitter have enabled feminist
campaigns to spread rapidly, mobilize, and build affective solidarity by shar-
ing experiences of discrimination and sexual violence (Bates, 2014; Mendes,
Ringrose, & Keller, 2019). However, social media have also become signif-
icant sites of sexist harassment, as feminist activists have been targeted and
threatened (Amnesty International, 2018; Cochrane, 2013; Jane, 2017).
The affordances of message forums are better suited to deeper discussion and
community-building, because the structure and slower pace of threads enable

greater reciprocity and reflexivity (Graham, Jackson, & Wright, 2016).
OSS is not only a message forum but also a hybrid media creation, merg-
ing the symbolic resources of Hollywood with the internet. Watson’s decision
to establish a feminist book group was consistent with her most well-known
persona in the field of global film entertainment. Her continued association
with Hermione Granger in the Harry Potter movies reinforced her image as a
“purely good character” (O’Donnell, 2017, p. 117), aligned with aspirational
values of “high achievement” (Mendick, Allen, Harvey, & Ahmed, 2018,
p. 156). However, Watson’s obvious institutional connections and elite celeb-
rity status might have jeopardized her claim to represent her OSS constitu-
ency. If Watson did not live up to the implicit promise to participate on equal
terms with the OSS user-audience, there was potential for disappointment
among those seeking interaction. Although online communities of this sort
are not without forms of leadership, they rarely feature fixed hierarchy and
centralized authority (Bruns, 2008). If celebrities’ ability to obtain political
capital is contingent on claims to represent user-audiences, how did Watson
make such claims? And how did user-audiences respond?
Fieldwork and Data
We use an online ethnographic approach to study how Watson made claims to
represent the OSS group across fields and platforms. The lead author (Watts)
joined the group in March 2016, reading the books selected for discussion,
occasionally posting messages, and monitoring online coverage of Watson
through daily Google News alerts. She collected and made notes on the
following content between January 2016 and January 2017: Watson’s 32
posts on the OSS forum; her interviews with feminist authors; her Goodreads
profile; her presentation of herself and OSS across Facebook, Twitter, and
Instagram; and references to Watson’s activism in online news and enter-
tainment media. These data were analyzed through open thematic coding to
assess how Watson presented her role in OSS and her relationship to user-au-
dience networks. The analysis included, for example, tagging references to
Watson’s UN role, language positioning her among OSS members, and state-
ments from Watson and OSS members inviting interaction. We use these data
to demonstrate how Watson performed three distinct representative claims,
while managing her proximity and distance from user-audiences.
We wanted to understand OSS members’ motivations for engaging
with the group, and their evaluations of Watson as a representative, with-
out relying only on accounts from the group’s most active members. The
18 Watts & Chadwick
22 participants—recruited through a message Watts posted on the forum—

included some of the most active members who had posted over 1,000 times
but also four members who had never posted at all. Participants (referenced
herein by pseudonym) ranged in age from 19 to 69 and were living in nine
countries across Europe, North America, and Central America. Interviews
were conducted by Watts through email, Goodreads private messages, and
Skype. Participants were sent nine questions; we draw on responses to the
following questions in this chapter:
1. Why did you want to join Our Shared Shelf?

2. Were you already following Emma Watson’s feminist activism before
(through HeForShe and/or through her social media)?
3. If so, what was it about Emma Watson and/or her activism that made you
want to get involved?
4. What do you do on OSS, and what do you most enjoy about being
part of it?
Watson’s Representative Claims
On launching Our Shared Shelf Watson told readers she would “post some
questions/quotes to get things started” and invite “prominent voices” to
“join the conversation,” a conversation she framed as an “open discussion
with and between you all” (Our Shared Shelf, 2016). However, Watson’s
visible engagement on the forum during the period of analysis was lim-
ited: she did not interact with other members, and her self-presentation
was guarded. To understand how Watson’s political capital derived from
claims to represent OSS therefore required going beyond the boundaries
of its message forum. We find that Watson used digital media more broadly
to perform three distinct claims to represent user-audience networks; we
term these “connected representative,” “ordinary member,” and “authentic
ambassador.”
Watson as Connected Representative

Watson’s self-presentation on OSS was predominantly professional and imper-
sonal. Her Goodreads profile featured a black and white headshot and sparse
personal information, her only listed interest being Our Shared Shelf. This
caution extended to her limited use of Goodreads’ affordances for sharing;
Watson did not rate the books she introduced to the group or write reviews
upon which others could comment. This was indicative of how Watson con-
structed her relationship with co-participants, a relationship framed around
responsibilities toward rather than rapport with others. Watson noted the
responsibility she felt to “figure out the next best thing to read” for a group
which had become “much more international than … expected—and much
bigger” (Watson, 2016d, 2016e). Acting as an educated facilitator, Watson
encouraged members to link books to political issues while maintaining her
distance by rarely sharing her own views. Introducing Margaret Atwood’s
The Handmaid’s Tale, for example, Watson (2017b) encouraged members
to think “beyond the tag” and “share our thoughts about how we think
its dystopian vision relates to the world of 2017.” In many respects, this
approach—in which Watson played an enabling role rather than seeking direct
mobilization—derives from an earlier period in the web’s development. This
is in stark contrast with the social media influencer model that has become
dominant in recent years.
Watson promised that she would be “harassing whoever I need to harass
to get questions answered” (Watson, 2016d). By interviewing feminist
authors on behalf of OSS, Watson demonstrated her growing political cap-
ital by connecting the group to her own elite networks. Some interviews
also afforded opportunities to represent OSS to broader audiences; Watson’s
interview with Persepolis author Marjane Satrapi, for example, was published
by Vogue. This interview, however, highlighted the tensions generated by
attempting to balance proximity and distance. Although Watson (2016g)
promised to ask “as many as I can,” she put only two member questions to
Satrapi during her conversation with the author. Responses on the forum sug-
gested members appreciated this personal style, praising the “genuine conver-
sation” and “unedited” exchange between people with a “real connection.”
It therefore appeared less important that Watson directly represent OSS’s
views to broader audiences than that OSS could gain a backstage glimpse of
the guarded star. This raises the question of whether Watson’s generally pro-
fessional, even impersonal self-presentation placed her at too great a distance
from members to be accepted as genuine.
Watson as Ordinary Member of OSS

Watson also used language in her forum posts to construct a very different
type of representative claim—one that positioned her among others as an
ordinary member with shared interests and experiences. Watson’s (2016b)
first book announcement struck a conversational tone; she told members
she was “reading it with a pen in hand” and making “a cup of peppermint
tea.” Watson’s announcement posts continued to give the impression she was
“learning and reading with” co-participants. She said that she was “excited”
to “read this book with you” and to “hear what you think” (Watson, 2016c,
2016e, 2016f).
20 Watts & Chadwick
Crucially, this positioned Watson as fellow learner rather than all-know-

ing authority, and seemingly flattened the hierarchy between her and OSS
members. Watson (2017a) used an uncharacteristically long and personal post
(to announce OSS would read Reni Eddo-Lodge’s Why I Am No Longer
Talking To White People About Race) as a means of addressing criticism that
her feminism was exclusive. Here Watson’s UN speech was no longer a source
of expertise, but the start of what she described as a “journey” and an “inter-
rogation of self.” She related this to each member’s “own journey,” telling
co-participants she was “looking forward to discussing” the book “in more
detail … soon.”
Watson benefited from this opportunity to perform her learning through
claims to ordinariness. Her post was remediated through online news and
entertainment media, as commentators praised her “acknowledgment” of
White privilege and “lesson in self-awareness” (Animashaun, 2018; Bradley,
2018; Canty, 2018; Kelly, 2018; Muller, 2018; Okolosie, 2018). Despite
this claim to be on a shared journey, in practice Watson’s direct engagement
with forum members was close to non-existent. Between January 2016 and
April 2017, Watson published 34 posts, 24 of which were announcements.
Although Watson’s (2016a) first-ever post reassured a member that “I’m here!
I am having the best time reading these discussion boards!,” of the eight posts
we coded as interactions, seven were made in the group’s first two weeks. Her
announcement posts received between 126 and 1,241 responses, suggesting
an appetite for interaction that might have quickly turned to disappointment.
Watson’s celebrity status was difficult to reconcile with claims to ordinary
participation in an online community. She was both a connected represen-
tative above the group and an ordinary member among co-participants. We
now discuss how Watson used social media to perform a third, more complex
claim to be an authentic ambassador for members across fields and platforms.
Paradoxically, these platforms—often associated with interactivity—enabled
Watson to perform engagement from a distance.
Watson as Authentic Ambassador

Watson was most visible as the group’s representative outside of OSS, where
she used social media to retain connection with members at a distance as their
authentic ambassador. This positioned Watson both within and above the
group, her celebrity capital and social media posts enabling her to represent
OSS to wider audiences. She did this directly by performing her engage-
ment through social media and indirectly as this content became remediated
through online news. By using social media and not the group forum itself to
perform representative claims, she maintained the distance from user-audience
networks that has traditionally been associated with celebrities of high status
(Marshall, 2014; Rojek, 2001).
Instagram was essential to performing this role. Sharing selfies with books
and reposting content from the group’s Instagram account, she broadened
her invitation to “let me know what you think” to her 50 million followers
(oursharedshelf, 2017). She used social media to create the opportunity for
members to feel they were “reading along” with her, in real time: she posted
a selfie with the first group selection (Gloria Steinem’s My Life on the Road),
asking followers “Who has their book?” (emmawatson, 2016a). Following
this, members began sharing their own OSS selfies (or “shelfies” as they
became known) to demonstrate their participation, and by January 2019,
#OurSharedShelf had been used in almost 24,000 Instagram posts. By stating
she could “literally see” these contributions, Watson (2016d) reinforced the
impression that she and other members were co-producing a campaign in and
beyond the OSS forum.
The attention Watson received from international news and entertain-
ment media sources supported her representative claim, as she connected
OSS to wider audiences. When Watson collaborated with the Books on The
Underground project in November 2016—leaving copies of a Maya Angelou
book selected for OSS in London stations—her Instagram video was viewed
over 4.2 million times and 64 news articles about her intervention were pub-
lished (emmawatson, 2016b). Watson documented her engagement with
feminist campaigns across social media, mediating, for example, her participa-
tion in the Women’s March in Washington DC in January 2017 on Facebook
(Emma Watson, 2017).
This claim was performed not only across platforms but also across fields,
as illustrated by her public reflections about her starring role in Disney’s 2017
remake of Beauty and the Beast. Watson claimed she had “turned down”
Cinderella because the lead character was not a “role model” (Frost, 2017),
instead crafting a backstory of “empowering defiance” for the character of
Belle in Beauty and the Beast (Furness, 2017). When her view that Belle
is a feminist role model was contested, Watson told Entertainment Weekly
(2017) she had shared these concerns and addressed them by “doing some
reading.” Watson even took OSS author Gloria Steinem to the film’s pre-
miere (MacKelden, 2017). This consistency “across all aspects of [her] life
and communications,” which Marwick (2013, p. 240) argued is key to per-
ceived authenticity, supported Watson’s claim to be ambassador for OSS in
her absences from the forum.
Thus, Watson performed three types of claim to represent members
of her online feminist book group and discussion forum. As connected
22 Watts & Chadwick
representative she foregrounded elite connections, while as ordinary member

she positioned herself as a fellow learner. By channeling her engagement with
OSS outside the forum, particularly on Instagram, Watson balanced prox-
imity and distance from user-audiences by acting as authentic ambassador
across fields. Although these claims co-existed in tension, Watson’s ability
to perform them demonstrates the volume and variety of her resources. But
was Watson accepted as a representative of an online community grounded
in co-participation, despite her elevated status? We now turn to our analysis
of interviews with a small sample of OSS members. We show that Watson’s
distance from members—both in terms of her engagement with them and her
elite connections—was in fact key to the broad acceptance of her claims to
represent feminism in the political field.
How did User-Audiences Evaluate Watson’s Representative

Claims?
Interviews with OSS members suggested the relationship between the group’s
celebrity founder and its user-audiences was complex. Our aim here is not to
make generalizations about the views of the entire membership of OSS, but
to explore the connections between Watson’s representative claims and how
the 22 members we interviewed described the celebrity and her relationship
with the group. This first requires some understanding of Watson’s role in
motivating these members to join. When asked what prompted them to join
OSS, participants’ responses suggest they were often (though not exclusively)
made aware of the group through content posted by or about Watson. Eight
noted seeing OSS on Watson’s social media, five noted prior engagement
with HeForShe, and five reported seeing online news articles about OSS and
Watson with two specifically noting her aforementioned collaboration with
Books on the Underground/Subway. In contrast, three mentioned the elec-
tion of Donald Trump as a prompt for joining OSS, whereas only two noted
finding the group through the Goodreads website itself.
Although most participants discovered OSS through Watson, it would be
a mistake to assume that members are motivated to join purely by the potential
for proximity to the celebrity. Eight participants noted Watson’s involvement
as a reason for joining, but members held multiple motivations for wanting to
do so; a love of reading (n = 12), looking for a community (11), and wanting
to learn (11) were mentioned more frequently. Participants also described an
identification with feminism (8), looking for discussion (5), wanting to take
action (4), and looking to teach others (2) as motivating factors. Our 22 par-
ticipants ranged from undying Watson fan to uninterested reader.
We argue that Watson’s use of digital media to represent OSS from a dis-
tance afforded her broad acceptance from these varied members. Interestingly,
those who had followed Watson’s journey most closely did not want to see
her engage more directly with the group. Her hands-off role was seen as
appropriate, and OSS not the platform to seek interaction. Alex, for example
(all names are pseudonyms), said that she sent Watson multiple letters but
was “comfortable” with her role in OSS, “posting about the new book and
that was more or less it.” By not intervening in discussions, Watson avoided
being seen to speak over rather than for members, behaving as if, in Alex’s
words, “I’m the big queen and I’m going to rule over every one of you!”
Rosa agreed that OSS was not the place for Watson’s opinions: “I like the
way she proposes books and thoughts of others, not presenting them as her
own philosophy.”
Watson’s celebrity capital was necessary to her acceptance as a represen-
tative, however, due to what she did with her status, when she gave “voice
to a lot of women that haven’t that choice” and used her “voice for some-
thing positive in the world” (Bianca; Maria). The participants who described
Watson as admirable or inspirational often based their views on Watson’s use
of fame to promote feminism (see Table 2.1). Watson’s representative claims
were accepted because she could “get more audience” for feminist issues,
bring “a huge (and certainly diverse) crowd of people” together, and make
“gender equality issues more accessible for the “every day” person” (Rosa;
Louise; Chloe). The scale of Watson’s celebrity capital was therefore essential
to its exchangeability, her ability to attract attention to OSS and promote its
values key to her acceptance by user-audience networks.
Table 2.1. Perceptions of Watson as a Representative
Watson is perceived as Number of participants

Admirable/inspirational 9
Serious (due to institutional links) 6
Serious (in comparison with other celebrities) 6
Relatable 6
Authentic (“genuine”, or doing things “for the right 4
reasons”)
Committed to the cause 4
Trustworthy 3
Knowledgeable 3
A role model for young women 2
24 Watts & Chadwick
Participants often discussed Watson by comparing her to “other celebri-

ties.” These comparisons revealed participants’ discomfort with associating
too closely with celebrities in general, but—and this is significant—not Watson
specifically. Rosa revealed her initial concern about listening to Watson “just
because she’s famous”—which Rosa said felt like “teenage behaviour”—but
changed her mind after “reading her posts and listening.” Similarly, Matthew
described himself as “wary of celebrity” but after “learning more of her life”
through “her media presence” concluded Watson was “as hardworking as she
is gifted and earned all she had.” Thus, Watson’s engagement at a distance
enabled her to gain acceptance as an exceptional type of political celebrity.
Intriguingly it was not only Watson’s spatial distance from members, but
also her social distance from them, which afforded this acceptance. Views
about Watson’s role as UN Women Goodwill Ambassador often underpinned
group members’ perceptions of her as admirable or inspiring, and behaving
in the “right way” for a celebrity who wishes to intervene in politics. Tricia,
for example, was “not interested in ping pong Twitter insults or threats” and
saw Watson as “taking the high road and going through formal channels …
i.e. the UN.” Watson’s official capacity at the UN also strengthened her claim
to be in a position to effect political change. For Alex, Watson’s ability to talk
“to Justin Trudeau and to so many people” meant, in a telling phrase, that it
was “a bit ridiculous to question her.”
This perceived appropriateness of Watson’s self-presentation also afforded
her acceptance by forum members as a “deserving” celebrity—a class-based
distinction Mendick et al. (2018) have argued has become central to how
young people understand meritocracy. For Alyssa, it was “nice to see a young
celebrity who’s not getting involved in scandals and drugs … actually doing
good productive work in the world.” Although Claudia praised other young
celebrities who “stand up,” Watson was contrasted positively as “more down
to earth and considerate” than those who are “loud and have a kind of ‘I
don’t give a shit’ vibe.” These comparisons benefited Watson: she was seen as
using her celebrity capital to “engage in issues that really matter and do some
good in the world—and not just for a PR stunt,” explained Chloe.
Crucially, however, the distance afforded by Watson’s elite institutional
connections and appropriate self-presentation did not preclude her from being
described as “relatable,” “genuine,” and “trustworthy.” Sophia described
Watson as “genuine” because in “every interview or article written about her
she has the same message shine through that makes you trust and believe in
her.” For Chloe it was admirable that Watson had not been “spoilt” by having
been “thrown into the public eye,” again supporting Mendick et al.’s argu-
ment that consistency matters because audiences assess whether celebrities
appear “changed” in a negative way by their wealth and fame (2018, p. 60).
Although Watson’s high celebrity capital therefore supported a claim to have
wide reach, exchanging this celebrity capital through acceptance as a political
representative is a fine balancing act with audiences at its core.
Digital Distance: Celebrity Power and User-Audience Networks

In this chapter, we have explained how Emma Watson used her celebrity cap-
ital and media platforms to construct claims to represent Our Shared Shelf,
the feminist book group and discussion forum she founded in 2016, and, as
a consequence, her claims to represent feminist ideas more broadly. Watson’s
celebrity status, and indeed her three claims to represent other members,
could potentially have been at odds with the aims and affordances of the
group. We have argued that representative claims are necessary because they
act as mechanisms through which celebrities attempt to exchange celebrity
capital for political capital. We demonstrate that the scale or magnitude of
celebrity capital alone is not sufficient to produce the kind of recognition
that is required to act in the political field. Although the OSS members
valued Watson’s elevated status, there was also some reluctance to associ-
ate with celebrity in general. By founding an online forum for conversa-
tion “with and between you all,” Watson carefully negotiated the balance
between proximity to, and distance from, the user-audience networks that
were key to legitimating her role as a political agent. But this balancing act
was a challenging one. Would participants seeking interaction with Watson
as a co-participant in an online forum be disappointed with her limited direct
engagement in practice? Would Watson’s proximity to members hinder her
efforts to obtain political recognition, due to the interconnection between
distance and status?
Although social media have previously been associated with interactivity
and reduced distance between celebrities and their audiences (Turner, 2014),
we found that social media enabled Watson to represent user-audience net-
works while retaining appropriate distance.
Watson’s posts on the OSS forum itself constructed two types of claim
that were challenging to reconcile: she drew on her status and connections
as the group’s connected representative, while also positioning herself as an
ordinary member of the group. Through her social media practices, Watson
performed a more complex claim to be an authentic ambassador for user-au-
dience networks, representing their broadly shared interests across fields
and platforms while rarely revealing details of her personal life. This enabled
Watson to keep her distance while retaining her role as a representative.
26 Watts & Chadwick
Indeed, we argue Watson’s balance of proximity and distance was a fun-

damental element of her acceptance as a representative of feminism. Watson
was perceived as a sufficiently serious representative in comparison with other
celebrities due to her connection to the UN and positive assessments of her
cautious self-presentation. Our analysis suggests that the management of
proximity and distance remains key to the maintenance of celebrity status—
and key to how celebrity capital can be exchanged for political capital.
References
Abidin, C. (2018). Internet celebrity. Bingley, UK: Emerald Publishing.
Amnesty International. (2018). #ToxicTwitter: Violence and abuse against women online.
London: Amnesty International Ltd. Retrieved from https://www.amnesty.org/
download/Documents/ACT3080702018ENGLISH.PDF.
Animashaun, D. (2018, January). Reni Eddo-Lodge & why she’s no longer talking to
white people about race. Konbini. Retrieved from http://www.konbini.com/ng/
entertainment/reni-eddo-lodge-shes-no-longer-talking-white-people-race/.
Bates, L. (2014). Everyday sexism. London, UK: Simon & Schuster.
BBC Newsbeat. (2014, July 8). Emma Watson announces UN Women Goodwill
Ambassador role. BBC Newsbeat. Retrieved from http://www.bbc.co.uk/newsbeat/
article/28205765/emma-watson-announces-un-women-goodwill-ambassador-role.
Bourdieu, P. (1987). What makes a social class? On the theoretical and practical existence
of groups. Berkeley Journal of Sociology, 32, 1–17.
Bourdieu, P. (1991). Language and symbolic power. Cambridge, UK: Polity Press.
Bradley, N. (2018, January 10). Emma Watson’s comments on “white feminism” are a
lesson in self-awareness & intersectionality. Bustle. Retrieved from https://www.bus-
tle.com/p/emma-watsons-comments-on-white-feminism-are-a-lesson-in-self-aware-
ness-intersectionality-7842357.
New York, NY: Peter Lang.
Canty, E. (2018, January 12). People called Emma Watson a ‘white feminist.’ Now, she
admits, they weren’t wrong. Upworthy. Retrieved from http://www.upworthy.com/
people-called-emma-watson-a-white-feminist-now-she-admits-they-weren-t-wrong.
Chadwick, A. (2017). The hybrid media system (2nd ed.). Oxford, UK: Oxford
University Press.
Chadwick, A., O’Loughlin, B., & Vaccari, C. (2017). Why people dual screen political
debates and why it matters for democratic engagement. Journal of Broadcasting &
Electronic Media, 61(2), 220–239.
Cochrane, K. (2013). All the rebel women: The rise of the fourth wave of feminism. London,
UK: Guardian Books.
Driessens, O. (2013). Celebrity capital: redefining celebrity using field theory. Theory and
Society, 42(5), 543–560.
emmawatson. (2016a). Who has their book? #OurSharedShelf #BookClub

#MyLifeOnTheRoad. (Instagram post) January 14. Retrieved from https://www.
instagram.com/p/BAhgjIIn3P1/?taken-by=emmawatson
emmawatson. (2016b). @booksontheunderground@oursharedshelf #Mom&Me&Mom.
(Instagram post) November 1. Retrieved from https://www.instagram.com/p/
BMSHByylwn5/
Emma Watson. (2017). #IStandWithPP why the fight for reproductive rights goes
on. (Facebook post) 21 January. Retrieved from https://www.facebook.com/
emmawatson/posts/1493263300692555
Entertainment Weekly. (2017, February 16). Emma Watson responds to claims that Belle
has Stockholm Syndrome. Entertainment Weekly. Retrieved from https://www.you-
tube.com/watch?v=51NxeWInaxI.
Forbes. (2017). The top-grossing film franchises. Forbes. Retrieved from https://www.
forbes.com/pictures/fimi45fhgkg/2-harry-potter/#1b0fbb0a38ed.
Frost, K. (2017, January 17). Emma Watson reveals why she turned down Cinderella
before accepting Belle role. ELLE. Retrieved from https://www.elle.com/uk/
life-and-culture/culture/news/a33458/emma-watson-turned-down-cinderella-
before-belle.
Furness, H. (2017, January 17). Emma Watson confirms she turned down Cinderella
before taking on ‘role model’ Belle. The Telegraph. Retrieved from https://www.
telegraph.co.uk/news/2017/01/17/emma-watson-confirms-turned-cinderella-
taking-role-model-belle/.
Graham, T., Jackson, D., & Wright, S. (2016). ‘We need to get together and make our-
selves heard’: everyday online spaces as incubators of political action. Information,
Communication & Society, 19(10), 1373–1389.
HeForShe. (2015). Emma Watson speech for HeForShe IMPACT 10x10x10 Program
at World Economic Forum 2015. Retrieved from https://www.youtube.com/
watch?v=oE28bb11GQs.
HeForShe. (2016). The HeForShe commitment. HeForShe. Retrieved from http://www.
heforshe.org/en/commit.
Horton, D., & Wohl, R. (1956). Mass communication and para-social interac-
tion: Observation on intimacy at a distance. Psychiatry, 19(3), 215–229.
Jane, E. A. (2017). Misogyny online: A short (and brutish) history. London, UK: SAGE.
Jenkins, H. (2006). Convergence culture: Where old and new media collide. New York,
NY: New York University Press.
Kelly, E. (2018, January 11). Emma Watson acknowledges her white privilege in open letter
about race and feminism. Metro.Retrieved from http://metro.co.uk/2018/01/11/
emma-watson-acknowledges-white-privilege-open-letter-race-feminism-7220955/.
MacKelden, A. (2017, February 28). Emma Watson took Gloria Steinem to see ‘Beauty
& The Beast’ to make sure the movie was feminist enough. Bustle. Retrieved from
https://www.bustle.com/p/emma-watson-took-gloria-steinem-to-see-beauty-the-
beast-to-make-sure-the-movie-was-feminist-enough-41275.
28 Watts & Chadwick
Marshall, P. D. (2014). Celebrity and power: Fame in contemporary culture.

Minneapolis: University of Minnesota Press.
Marwick, A. (2013). Status update: Celebrity, publicity, and branding in the social media
age. New Haven, CT: Yale University Press.
Marwick, A., & boyd, d. (2011). To see and be seen: Celebrity practice on Twitter.
Convergence, 17(2), 139–158.
Mendes, K., Ringrose, J., & Keller, J. (2019). Digital feminist activism: Girls and women
fight back against rape culture. New York, NY: Oxford University Press.
Mendick, H., Allen, K., Harvey, L., & Ahmed, A. (2018). Celebrity, aspiration and con-
temporary youth. London, UK: Bloomsbury.
Muller, M. G. (2018, January 9). Emma Watson addresses her white privilege and ‘white
feminism’ in letter to her book club. W Magazine. Retrieved from https://www.
wmagazine.com/story/emma-watson-white-privilege-feminism.
Nichols, L. (2014, September 24). UN Women sees major social spike after Emma Watson
speech. PR Week. Retrieved from https://www.prweek.com/article/1313911/
un-women-sees-major-social-spike-emma-watson-speech.
O’Donnell, K. M. (2017). Celebrities and conflicting notions of modern feminist embod-
iment. In J. Raphael & C. Lam, (Eds.), Becoming brands: Celebrity, activism and
politics (pp. 110–121). Toronto, Canada: WaterHill Publishing.
Okolosie, L. (2018, January 20). Emma Watson’s willingness to face the truth about
race is refreshing. The Guardian. Retrieved from https://www.theguardian.com/
commentisfree/2018/jan/10/emma-watson-truth-race-white.
Our Shared Shelf. (2016). Our Shared Shelf. Goodreads. Retrieved from https://www.
goodreads.com/group/show/179584-our-shared-shelf.
Oursharedshelf. (2017, February 27). Please let me know what you think of the
book, share your photos and be sure to tag #Oursharedshelf!! Happy reading
❤🐺#womanwhorunwiththewolves #clarissapinkolaestés #MarchApril. (Instagram
post). Available at: https://www.instagram.com/p/BRBlyehFZQ9/.
Rojek, C. (2001). Celebrity. London, UK: Reaktion Books.
Saward, M. (2010). The representative claim. Oxford, UK: Oxford University Press.
Turner, G. (2014). Understanding celebrity (2nd ed.). London, UK: Sage Publications.
UN Women. (2014, September 20). Emma Watson: Gender equality is your issue
too. Retrieved from http://www.unwomen.org/en/news/stories/2014/9/
emma-watson-gender-equality-is-your-issue-too.
Watson, E. (2016a, January 8). I’m here! I am having the best time reading these dis-
cussion boards! Goodreads. Retrieved from https://www.goodreads.com/topic/
show/17961433-where-is-emma.
Watson, E. (2016b, January 8). First book!! My life on the road, by Gloria Steinem. Goodreads.
Retrieved from https://www.goodreads.com/topic/show/17965881-first-book-
my-life-on-the-road-by-gloria-steinem.
Watson, E. (2016c, February 2). Second book!! The color purple, by Alice
Walker. Goodreads. Retrieved from https://www.goodreads.com/topic/
show/18005443-second-book-the-color-purple-by-alice-walker.
Watson, E. (2016d, February 4). 100,000 members! Goodreads. Retrieved from https://
www.goodreads.com/topic/show/18006758-100-000-members.
Watson, E. (2016e, April 23). May book!! The Argonauts, by Maggie
Nelson. Goodreads. Retrieved from https://www.goodreads.com/topic/
show/18093486-may-book-the-argonauts-by-maggie-nelson.
Watson, E. (2016f, May 16). June book!! Persepolis, by Marjane Satrapi. Goodreads.
Retrieved from https://www.goodreads.com/topic/show/18114001-june-book-
persepolis-by-marjane-satrapi.
Watson, E. (2016g, June 7). Questions for Marjane Satrapi! Goodreads. Retrieved from https://
www.goodreads.com/topic/show/18141113-questions-for-marjane-satrapi.
Watson, E. (2017a, January 2). First book of 2018! Why I’m no longer talking to white
people about race by Reni Eddo-Lodge. Goodreads. Retrieved from https://www.
goodreads.com/topic/show/19152741-first-book-of-2018-why-i-m-no-longer-
talking-to-white-people-about-race#comment_174809853.
Watson, E. (2017b, April 17). May/June book! The Handmaid’s Tale. Goodreads. Retrieved
from https://www.goodreads.com/topic/show/18573966-may-june-book-the-
handmaid-s-tale#comment_id_165255055.
3.
L8r H8r: Commoditized Privacy,
Influencer Wars, and Productive
Disorder in the Influencer Industry
Crystal Abidin
Since their earliest commercial endeavors on blogs such as Blogspot or

LiveJournal and social networking sites such as Friendster or MySpace, ordi-
nary people who become internet famous have established themselves as
microcelebrities who utilize technology to brand themselves as authentic and
famous to a niche group of followers (Senft, 2008). A more sophisticated
incarnation of microcelebrities is Influencers—a highly viable, systematic, and
professional form of microcelebrity whose careers operate on ideogeograph-
ical-specific and platform-specific ecologies of attention, aesthetics, affects,
social ties, identities, and commerce across various social media, and between
mainstream and amateur media industries. Influencers make most of their
money from advertorials on their digital estates, as well as brand ambassador-
ships and appearances at physical events (Abidin, 2015).
In the last decade, the Influencer industry has become increasingly pro-
fessionalized and saturated as people attain digital literacies at a younger age
and find lower barriers to entry, digital affordances of platforms become more
user-friendly and universally accessible, and pioneering Influencers demon-
strate success and model scripts that aspiring Influencers can follow. Various
social media have emerged with dominant tropes and prominent users; for
instance, visually oriented Instagram has become the gold standard for pub-
lishing highly curated and congruent feeds featuring immaculate snapshots of
Influencers’ pristine lifestyles.
Some Influencers have begun to break away from the picture perfect mold
of Instagram and the stasis of their rank in the Influencer industry. A bold
group of Influencers is seeking new followers by commodifying their privacy
32 Crystal Abidin
to chase drama and controversy, garner negative attention for themselves or

others, display authenticity claims, engage in Influencer wars, and play with
ideas around productive disorder. I will call such users controversy-seeking
Influencers. Their hidden agenda can explain why, although some Influencers
struggle with haters and the backlash associated with the job (Abidin, 2013),
others invite and then dismiss such criticism with the maxim “later, hater” or
“l8r h8r.” This chapter looks at how Influencers negotiate between partici-
pating in negative attention rituals and relying on web amnesia to grow their
follower base.
The data in this chapter is constituted from two field sites. The contextual
knowledge of the Influencer industry and examples drawn from Influencers in
Singapore are informed by my ongoing anthropological fieldwork with blogs-
hop and Influencer cultures since 2008. This comprised participant observa-
tion among social media celebrities, their family and friends, their backend
staff, their management agencies, their clients and sponsors, and their fol-
lowers and haters in the flesh; personal interviews with the abovementioned
groups; content analysis of Influencers’ active digital estates; web archaeology
into Influencers’ abandoned digital estates; and archival research on press
mentions and populist discourses on Influencers. The in-depth case studies
of Influencer-related commotions and events are based on my research on
the attention and aesthetic economy of various social media since 2014. This
comprised archiving and analyzing in real time events such as global tragedies
via trending hashtags, vernacular virality and memes, Influencer scandals, and
changes on platforms including Blogspot, Wordpress, Instagram, Snapchat,
Twitter, and Facebook. I focus on three such Influencers scandals between
2012 and 2014. But to understand how Influencers come to play with neg-
ative attention rituals, we must first appreciate how Influencers at different
ranks and stages of their careers experience differentiated privileges when
attempting to commoditize their privacy.
Commoditized Privacy: A Lifecycle

The personae of Influencers are premised upon sharing selected aspects of
their lives that are usually personal and publicly inaccessible. Therefore, pri-
vacy becomes a commodity that is manipulated and performed to advance
their careers. At low-status, privacy is deemed a necessary sacrifice for career
growth until it is distinguished as Influencer persona privacy and non-Influ-
encer persona privacy. At mid-status, persona privacy is a calibrated perfor-
mance to increase readership. At high-status, all privacy becomes a privilege
with intrinsic value.
L8r H8r 33
Low Status
In the early stages, Influencers have not yet developed Influencer personae
nor distinguished them from non-Influencer identities. They conceptualize
privacy as a personal quality based on their most private, non-commercial
identities and desire to preserve it. However, success in the Influencer indus-
try is measured by the volume of oneʼs viewer traffic, and Influencers struggle
between preserving their privacy but settling for low readership, or sacrificing
their privacy and acquiring high readership.
Cassandra, who had 1,200 daily blog views when I interviewed her at
a very early point of her career, is unwilling to sacrifice too much privacy.
She has stalled her career by intentionally remaining “low profile” and only
blogs about things she feels “will not attract too much attention.” In con-
trast, Natasha, who had once blogged about her experiences of underage sex
to 30,000 viewers, feels she no longer “owns privacy”—not because of her
blogposts’ content, but her extensive popularity. As a high-status Influencer,
Natasha deems this a “trade off” for her career.
Trading off between privacy and readership is confined primarily to
early stages of careers when Influencers have low-status. As they distinguish
Influencer personae from their non-Influencer identities, privacy becomes
conceptualized as two layers: one for the commercial persona, and one for
the personal identity. Commercial persona privacy is sacrificed, and personal
identity remains intact.
Mid Status
After developing an Influencer persona, mid-status Influencers are concerned
with increasing their readership. Many capture attention by turning usually
private events into a public performance. Privacy is manipulated into a public
staging, to captivate an audience in search of spectacles (Kitzmann, 2004).
As the most taboo, sex captures the largest audience. So-called “Leaked” sex
videos, “staged” domestic violence, and breakup “tell-all” exposés are inten-
tionally produced to bait attention. Holly states on her blog that her “leaked”
sex video “needed the chance to get your attention and sink in.” Like many
Influencers, she intentionally stages intimate moments for voyeuristic con-
sumption as a business strategy (Abidin, 2017).
Some mid-status Influencers worry about nuclear family members read-
ing their blogs when they are staging privacy. Influencers are generally com-
fortable with personal friends and romantic partners reading their blogs; the
insecurity is because nuclear family members—who hold intimate knowledge
of an Influencerʼs most private personal identity—potentially threaten the
34 Crystal Abidin
congruence of the constructed narratives (Daniel & Knudsen, 1995). For

instance, Jayne was “pretty okay” about blogging her “private life” until her
older brother found her blog. He disagreed with some of her self-presentation
and began to police her blog content, causing her much frustration. Belinda,
however, “feels safe” because her mother is “not computer savvy” and unlikely
to read her blog. Family is less of a concern for established Influencers who do
not feel the need to stage privacy to sustain followersʼ attention.
High Status
Once Influencers have captured a sizable following, withholding information
about their private lives acquires commodity value, because the mystique over
what is not displayed makes followers curious; the less revealed, the more
enticed followers are. Marianne notes that Anna can “afford to be private
about her life now [because] she is more successful.” Although Anna used to
publish raw pictures about life behind-the-scenes, her blogposts are now infre-
quent and more polished. On her social media feeds, followers leave hundreds
of comments asking about her relationship. Tracy remarks that high-status
Influencers are “classy Influencers” who do not need to “push themselves all
the time, [because] people will still want to know about [them].”
Alberoni (2007) noted that the elite class experiences less observability
and more secrecy. Papacharissi (2010) has conceptualized privacy as a similarly
privileged commodity, and in this case only among high-status Influencers,
whose non-disclosure solicits as much attention as their disclosure of informa-
tion. Influencers who have attained a particular standard and traction among
their followers can play with privacy as attention bait to stimulate desire and
excitement. Influencers pride themselves on being “ordinary people”; they
are accessible to followers and more relatable than mainstream celebrities
(Turner, 2010). Losing this status would jeopardize their credibility, so it is
paramount that high-status Influencers carefully negotiate a balance between
revealing and concealing their private lives.
Privacy for Profit
Turner argued that public figures become celebrities at “the point at which
media interest in their activities is transferred from reporting on their pub-
lic role … to investigating the details of their private lives” (2014, p. 8).
Geraghty (2007, p. 100–101) similarly noted that this form of “star-as-celeb-
rity” comprises attention focused on an individual’s “private life” irrespective
of their actual career or public personae. For Influencers, however, the private
and the public often overlap ambiguously and strategically masquerade as
L8r H8r 35
the other to bait followersʼ attention. As a form of “lifestreaming” (Marwick,

2013, p. 207), Influencersʼ conceptions of the private and the public under-
score Warnerʼs (2002, p. 414) third category of “publics” as a status that
“comes into being” through being broadcast, circulated, and widely publi-
cized. Marwick (2013) clarified Warnerʼs argument by stating that what mat-
ters in publicness is the intentional dissemination of information, rather than
the information simply being publicly available.
In fact, it is their very private lives that constitute their public personae, as
navigated via mechanisms of “presentational culture” (Marshall, 2010, p. 45)
afforded by social media technologies. In other words, privacy is no longer
personal seclusion in which one is free from public attention. Rather, it is
manipulated into a commodity for gain, and it differs across the Influencer
status spectrum.
Attention Events and Rituals

Goldhaber (1997) asserted that the scarcity of attention has generated what he
called “the attention economy.” He argued that “economies are governed by
what is scarce,” and that as abundant, overflowing information drowns us, we
must distinguish ourselves from the crowd. Goldhaber said that commanding
attention required originality, transparency, and the ability to convert atten-
tion into other resources and currencies. Davenport and Beck (2001, p. 2)
later added that although “capital, labor, information, and knowledge are
all in plentiful supply,” “human attention” is in shortage. They developed
three pairs of attention types: voluntary/captive, wherein one gives attention
by choice or not; attractive/aversive, wherein one gives attention for gains
or to avoid loss; and front-of-mind/back-of-mind, wherein one gives atten-
tion explicitly and consciously or out of habit (2001). Influencers usually
command a passive form of voluntary, attractive, and back-of-mind atten-
tion. However, the controversy-seeking Influencers I discuss here engage in
spectacle-like practices to generate an active form of captive, aversive, and
front-of-mind attention to recapture the foci of existing followers and attract
new ones.
The Influencer wars and negative attention rituals I discuss show how
Influencers convert bad publicity, self-shaming practices, and hating into
attention, which in return expands their follower traffic and increases the value
of the advertorial exposure they can provide. Influencer wars and negative
attention rituals are spectacles in that they are visually dominated with sym-
bolic codes of a “certain size and grandeur” (MacAloon, 1984, p. 243), and
serve as a “focal point of consciousness” and “means of unification” (Debord,
36 Crystal Abidin
2002, p. 6) in a social group. Boorstin (1961) called the orchestrated spec-
tacles I observe “pseudo-events”: news that is staged, executed for the mere
purpose of creating newsworthy content, bears an ambiguous representation
of the reality of events, and most crucially, becomes a self-fulfilling prophecy.
Debord similarly emphasized the false consciousness generated by specta-
cles that “ai[m]‌at nothing other than [themselves]” (2002, p. 7). With the
Influencers in this chapter, this is especially the case because the spectacles are
often merely exaggerated and dramatized accounts of and reactions toward
the mundane; Turner (2014, p. 92–93) termed this an “explosion of the
ordinary” that is mined as seemingly authentic and dedicated representations
despite actually being calculated productions of entertainment, or a “demotic
turn” in which (micro)celebrity culture is enabled by digital technology to be
increasingly ordinary although not necessarily increasingly democratic.
Anthropologically, these spectacular practices bear some semblance to
what Turner (1974, p. 33, 37) termed “social dramas”—“public episodes
of tensional irruption” in which conflict arises from “aharmonic” or “dis-
harmonic” processes. Social dramas are also concerned with the cohesion
and conflict within a social group. They can be productive to a group when
the conflict foregrounds the usually negligible “customs and habits of daily
intercourse,” causing people to “take sides in terms of deeply entrenched
moral imperatives and constraints, often against their own personal prefer-
ences” (p.35). Turner (p. 37–43) outlined four main phases of social dra-
mas: 1) “overt breach or deliberate nonfulfillment” of “norm-governed social
relations”; 2) escalation of the crisis causing a reordering of social relations;
3) redressive action initiated by “representative members of the disturbed
social system”; and 4) “reintegration of the disturbed social group” or “the
social recognition and legitimization of irreparable schism between the con-
testing parties.”
Influencer Wars: Three Case Studies

Influencer wars are short lived but intense events in which Influencers engage
in heated disputes with competitors through controversial claims to gener-
ate publicity for themselves. Similar clashes have been noted on YouTube
as “flame wars” in which “a flurry of video posts clusters around an internal
‘controversy’ or an antagonistic debate between one or more YouTubers”
(Burgess & Green, 2009, p. 97). Although Burgess and Green describe
YouTube flame wars as a ludic event that is spontaneous, undirected, and even
playful, the Influencer wars are deliberate publicity attempts. Through exag-
gerated and highly sensationalized accounts, Influencers stimulate widespread
L8r H8r 37
interest beyond their regular following, inviting other Influencers and their
followers to comment on the issue. The commotion produces a short time
in which Influencers can capitalize on general curiosity by producing insider
accounts of the controversy, joining camps (polarized supporters of opposing
parties in the dispute), or making provocative statements in order to join in
the Influencer war. This event disrupts the equilibrium of Influencersʼ relative
stable follower traffic, in which they can wrestle for attention, create publicity
for themselves, and increase their followings.
Constructing three genres of Influencer wars as case studies, I turn to
Influencersʼ engagements with status claims, authenticating appearance, and
tell-all exposés.
Status Claims
In July 2014, Influencer Eunice Annabel posted a picture of her manage-
mentʼs annual event, comprising a group of Influencers with the caption “So
you wanna be on top?”—a quote from the TV program Americaʼs Next Top
Model. She changed her Instagram profile title from “blogger” to “celeb-
rity blogger.” This was understandable and perhaps justified given her recent
movie and television appearances, and endorsement deals with various cos-
metics brands. She was a regular on magazines and newspapers and continu-
ously received good publicity from the press. Although she had been a child
actress, her recent prolific appearances in the mainstream media was her for-
mal crossover into the entertainment industry after having established herself
as an Influencer.
This angered a rival Influencer from a competing firm, Xiaxue, who
published a series of Instagram posts cryptically and directly criticizing
Eunice Annabel. This eventuated in two camps, supporting Eunice Annabel
or Xiaxue, cross-posting cryptic captions and critical statements of support
across various social media feeds. The Influencer war mostly occurred on
Instagram and Twitter, although several Influencers also published opinion
pieces about the incident on their blogs. This was widely dubbed “Xiaxue
vs. Eunice Annabel” by the mainstream media, with heated discussions and
follower camps breaking out on popular online forums and online news
outlets.
At stake was what constitutes “celebrity,” whether it can be achieved or
ascribed, and who was entitled to use the label. There was no formal resolu-
tion; both camps generated relatively equal amounts of support and hating.
However, after the commotion passed, Eunice Annabel edited her Instagram
biography again, removing the “celebrity” title.
38 Crystal Abidin
Authenticating Appearance
Influencer Seline has been accused of Photoshopping her photos since she
began blogging in 2005. She has refuted these claims in some instances, but
ignored others. Unedited photographs of Seline are widely circulating on the
Internet, and several threads on forums and blogposts are dedicated to expos-
ing her Photoshopping antics.
In July 2012, however, a relatively low-profile Influencer, Jermaine,
published a blogpost collating several of these active discussions, in a bid
to call out Selineʼs edited images. This blogpost circulated widely and was
cross-posted onto several social media platforms and online forums. The
post featured a string of flickering GIFs to demonstrate how much Seline
had doctored her images. Jermaine filtered through several forums, public
Facebook albums, and blogposts to compare and contrast Selineʼs before and
after images. Although it is widely known that Influencers use photo-enhanc-
ing applications, Influencers who do not disclose or who deny this practice
receive criticism from their counterparts. In Selineʼs case, the long-standing
and extensive doctoring of her images with no disclaimers thrust her into an
Influencer war for not being truthful about her self-representation.
“Tell-All” Exposés
In December 2013, Influencer Cassie published an Instagram photo of her-
self sitting on a man’s lap. Although this is not an unusual sight on her feed,
the deliberately hazy image featured a man who was not her then boyfriend
(who was well known among Cassieʼs followers). The image of this new man
was widely circulated, creating much gossip among followers, until a handful
of Influencers published social media posts identifying him. He was allegedly
a romantic interest of one of Cassieʼs best friends, and it was speculated the
two had been exchanging intimate correspondence despite Cassieʼs current
relationship. Cassieʼs best friend soon published a blogpost detailing what she
termed her “betrayal” and “hurt.”
In response, Fern was among the first Influencers to publish an exposé
of the issue, revealing that Cassieʼs mystery new boyfriend was one of her
ex-boyfriends. Fern wrote a lengthy blogpost entitled “Girlfriend code,”
arguing that ex-boyfriendsʼ best friends and best friendsʼ ex-boyfriends are
“strictly out of bounds” in the dating game. She also drafted several other
codes of “femininity” detailing the relationship boundaries she felt “girls”
could or could not transgress among each other. Many other Influencers and
followers published similar sentiments on social media platforms and blogs
calling for “sisters before misters” and “bros before hoes.”
L8r H8r 39
Although Cassie came out to clarify that she had already broken up with
her previous boyfriend a week before the incident, followers charged her for
not having “declared” or “announced” this publicly before posting the “inti-
mate” picture. Many Influencers also weighed in and chided her for dat-
ing again “so soon after the break up” and for having relations with a man
of whom her best friend was fond. Interestingly, most of the focus was on
Cassieʼs alleged “promiscuity,” with little discussion of the man’s behavior.
Cassie soon responded with what she termed a “heartfelt post,” bearing con-
notations of regret and hints of apologies. She also expressed surprise at how
quickly her Instagram photo went viral. However, the overarching discourse
about her “transgression” that was popularized by Fernʼs exposé and parroted
by others overshadowed Cassieʼs attempts at redemption.
Productive Disorder
Although it is tempting to brand such spats as mundane or trivial, and gloss
over them as mere gossip mongering, Influencer wars are actually a ritual of
disorder affecting everyday practices (Malefyt & Morais, 2012). Staging wars
and smear campaigns against competitors is a productive form of disorder
through which Influencers wrestle for followers’ attention and renegotiate
viewer traffic. Influencer wars generate captive, aversive, and front-of-mind
attention (Davenport & Beck, 2001) which entices new followers to observe
the confrontation and join a camp while strengthening existing followers’
allegiance.
Influencer wars such as status claims, authenticating appearance, and tell-
all exposés follow the cycle of social drama outlined by Turner (1974). In
each of these, an Influencer accuses another of committing a breach by using
a status-elevating title already claimed by a higher profiled Influencer, by
being dishonest about the use of photo-enhancing software, or by apparently
inappropriate dating behavior. Generating controversy in the industry gener-
ates hype or a frenzy of activity, in which the Influencer hierarchy’s stasis is
disrupted. Despite the apparent frivolity of things, these topics can command
attention and attract (good and bad) publicity, and function to appropriate
drama and controversy for individual Influencers’ gain.
In Influencer wars, the peak of the drama is the escalation, during which
the accuser produces a string of highly emotive and persuasive accounts
to convince fellow Influencers and followers of the accusedʼs wrongdoing,
resulting in a frenzy of users breaking into camps in support of one party and
a proliferation of attacking/defensive accounts from each camp. Low-profile
Influencers may seek the attention of passersby by capitalizing on this sense
40 Crystal Abidin
of disorder, attempting to produce side commentaries, personal editorials, or

mini (and often sloppy) exposés of their own promising previously unseen
information from behind the scenes as an insider—in summary, by producing
clickbait (Blom & Hansen, 2015). This creates publicity for themselves and
intensifies the exposure for their social media platforms through redirected
click-throughs.
As an attempt toward redressive action, Eunice Annabel omitted the title
celebrity blogger, and Cassie wrote a clarification blogpost and removed the
photograph from Instagram. However, Seline did not respond to the accu-
sations apart from a few cryptic and seemingly passive-aggressive statements
on her blog, suggesting that haters will always be “attracted to drama” and
are “not worth [her] time”. Engaging in wars, or responding if one happens
to be dragged in, is not always a viable option. Some Influencers stay away
from drama, save for the occasional cryptic one-liners (ironically) signifying
their disregard of haters and disengagement with the commotion. Others are
ambivalent and may comment only to refute allegations. Still others feel that
Influencer-warring is an inevitable element of their industry. Although some
Influencers appear more hesitant than others to speak up, almost all keep up
with breaking news and new scandals around the clock.
In the reintegration process, a new stasis is constructed in which alliances
among Influencers are reformed, and allegiances to the accuser and accused
that were publicly declared during the escalation process are publicly rein-
stated. Lines between each camp are made more defined. More crucially, fol-
lower traffic would have substantially increased for the accuser, the accused,
and the most vocal supporters within each camp, until the next Influencer war
breaks out to wrest attention away from the temporarily static hierarchy again.
Hating
Hating as a practice and vernacular concept among Influencers and their
followers warrants a brief discussion. As noted above, Burgess and Green
(2009) considered flame wars on YouTube an internal controversy or antag-
onistic debate among YouTubers manifesting as a high volume of video posts
within a short span of time. However, I want to focus on hating as a practice
among followers toward Influencers that may occur in peaks and troughs (as
in Influencer wars or negative attention rituals) or as an ongoing background
reaction to the voluntary, attractive, and back-of-mind attention (Davenport
& Beck, 2001) that Influencers elicit. In existing scholarship on the attention
economy, hating most closely resembles trolling. In her study of subcultural
L8r H8r 41
trolling practices, Phillips (2015, p. 15) defined “troll responses as those that
ʻfish for flames,ʼ ʻflamesʼ indicating an incensed response.”
Although Burgess and Green defined haters as “negative and often per-
sonally offensive commenters” (2009, p. 96), many Influencers I interviewed
perceive unanimous agreement in their industry that hating can sometimes
occur “just for the sake of it.” Influencers felt that hating comments were not
merely “harsh criticism,” but deliberately unproductive, hostile, and mali-
cious to generate ill will. Similarly, in her study of the term “troll” in Usenet
group rec.equestrian, Hardacker defined a troll as a person “whose real inten-
tion(s) is/are to cause disruption and/or to trigger or exacerbate conflict
for the purposes of their own amusement” (2010, p. 237). The extent and
momentum of hating generated by controversy-seeking Influencers could be
attributed to the fact that their spectacles accord with Birdʼs (2003) observa-
tion that long-lasting scandals generally dramatize and skirt the boundaries
of moral codes, invite judgment from followers, allow followers to engage in
dialogue such as in supporter and hater camps, appeal to emotions as human
interest stories, and are excessive to the point that followers are able to dis-
tance themselves from Influencers as violators.
Phillips observed that some early scholarship on trolling focused on
“effects-based definitions” (2015, p. 17), in which the practice is premised
on deception. However, she views trolling as a subculture “marked by a set
of unifying linguistic and behavioral practices” (p. 17), and that trolls are
motivated by “lulz,” an “unsympathetic, ambiguous laughter” in which trolls
“reve[l]‌in the misfortune” of those they dislike (p. 24). Contrary to popular
sentiment among followers I have interviewed that hating is “frivolous stuff,”
“just for fun,” and “has no effect in ʻthe real worldʼ,” haters and their hating
are valuable to Influencers in that they ultimately comprise follower traffic
and help raise awareness of and interest in the Influencer.
Hating accusations cannot always be verified and are often shrouded in
rumors and fictives (e.g., “I heard from a friend of a friend,” “According to
this unnamed source”). However, they can galvanize extensive support or
disregard for Influencers, as evidenced in the Influencer wars and negative
attention rituals evidenced above. Following from Phillipʼs (2015) analysis of
systemic subcultural trolling behaviors and drawing from my personal inter-
views among a small pool of followers (and haters), I summarize why some
followers engage in hating as a vernacular practice. Through a close coding of
my personal interviews with followers, I identified five prevalent discourses of
hating: counter-normativity, non-news, manufacturism, sensationalism, and
temporality.
42 Crystal Abidin
Haters repudiate controversy-seeking Influencers for being counter-nor-

mative and straying from the “mainstream” crowd. These Influencers are
chided for attracting “too much attention” to themselves:
Everything she does is just “me, me, me,” itʼs damn annoying lah … she is sooo
AA [“attract attention”—vernacular abbreviation referring to a person who war-
rants unnecessary attention]
Some of [the Influencers] are high profile for good things, like their achieve-
ments?… But [name of Influencer] is just always in the news for no reason …
everything also talk to reporters …
Although Influencers frequently headline newspapers and magazines, haters

highlight that much coverage of controversy-seeking Influencers is merely
frivolous and trivial gossip (i.e. Influencer spats and plastic surgeries). These
are occasionally labeled “first world problems,” after an Internet meme con-
noting that the exaggeration and disproportionate self-pity over very minor
frustrations are luxuries that only well-off peoples can afford. Many haters
reference major world events, such as wars and natural disasters, occurring as
Influencers dominate the national imaginary, to underscore a disproportion-
ate amount of publicity accorded to “non-news:”
I think itʼs damn lame because like, the front page news is about some stupid
bloggers fighting … or [having a] Twitter war, but itʼs not really news news like
people dying or what …
The third type of hating discourse focuses on the manufactured nature of

Influencersʼ controversies and gripes. These usually feature Internet users
complaining about Influencers who stage incidents of little substance, such as
if Influencers decried shaming incidents that followers did not feel breach any
moral code—for instance, Influencers who fail to mobilize weaponized shame
due to failing to understand vernacular shame, resulting in receiving reflexive
shame from followers:
… haiyah you know they say until like they [are] damn tragic, but who knows?…
maybe they all pakat pakat [conspire in secret] then come out to [create] drama
… itʼs always like that one
Haters also decry the sensationalist nature of Influencersʼ “antics,” citing

actions and statements getting blown out of proportion and coming across as
melodramatic and exaggerated. One hater mentions Cassieʼs “hazy Instagram
photo” discussed earlier, ridiculing how merely being photographed sitting
on a manʼs lap can “blow up” and invite insinuations that an Influencer “is a
slut” or “sleeps around.” Others observe that Influencer wars can break out
L8r H8r 43
as soon as one Influencer (mis)interprets anotherʼs “vague” Tweet as a smear

campaign against oneself despite no confirmation:
… who really knows [what vague Tweets refer to]? They are all so PA [Passive
Aggressive] … every small thing also make until so drama[tic]… like the situ-
ation is actually very small, but they can talk and hype until itʼs damn big deal
Lastly, haters deride the temporality and transience of Influencer drama, dis-
pelling the necessary effort to keep up with every single incident. Influencer
wars and negative attention rituals are constantly attempted by Influencers,
with attempts co-occurring and wrestling for followersʼ attention, result-
ing in attention fatigue. However, only some become recognized as actual
Influencer wars and negative rituals, replacing the stasis of voluntary, attrac-
tive, back-of-mind attention with captive, aversive, front-of-mind attention
(Davenport & Beck, 2001):
… after a while I was like, I give up, because the trends keep changing and
there is always a new [incident]… and they are all mostly the same just repeating
repeating repeating …
as soon as you [have been up-to-date] with one [incident], another one will
pop up …
Despite their denouncement of Influencersʼ controversy-seeking practices

(counter-normativity, non-news, manufacturism, sensationalism, and tempo-
rality), haters are still generally active and creative in their hating practices,
constituting a form of productive disorder for Influencers through increased
interest and traffic. In fact, haters and hating are so prevalent and effective that
laws have been enacted in response to Influencersʼ concerns over their safety,
reputation, and intellectual property rights: the Protection of Harassment
Act (November 2014) allows Internet users to be guarded from others who
cause them alarm, distress and abuse, including harassment, fear of provoca-
tion of violence, threats, and unlawful stalking. Influencer Xiaxue used this
act in January 2015 to obtain a Protection Order against Internet vigilante
group, SMRT Feedback (Ltd). She cited fear for the safety of her toddler and
husband, given that her personal information, including address and contact
number, was published on the SMRT Feedback (Ltd) Facebook page.
However, anonymous users on popular local forums speculated that this
move was merely Xiaxueʼs bid to silence haters, given that she has publicly
announced her (and her toddlerʼs) whereabouts on social media for years and
that much of her personal information is voluntarily archived on her blog.
We are beginning to observe something akin to Debordʼs notion of a cyclical
spectacle that “aims at nothing other than itself” (2002, p. 7), or Boorstinʼs
44 Crystal Abidin
(1961) pseudo-event, in which the generated news staged by controver-

sy-seeking Influencers, the solicited reaction from followers and haters, and
controversy-seeking Influencersʼ response to the hating form a feedback loop
that amplifies the synthetic novelty of self-shaming, in a self-fulfilling proph-
ecy that continually generates attention for Influencers. This is also evidenced
in the observation that an increasing number of Influencers are dedicating
entire blogposts addressing their haters despite being (recently) unprovoked.
Such blogposts are situated to aggravate haters and incite more hating, and
thus publicity, toward the Influencer.
On the whole, many of the Influencers to whom I spoke claim they try
their best to distance themselves from Influencer wars and negative attention
rituals. The majority acknowledges that controversy-seeking Influencers are
brave to engage in these provocative attention-gathering tactics, and that not
everyone is able to stomach the judgment and criticism that undoubtedly
accompany their actions. The attention-garnering strategies are often spectac-
ular, scandalous, and occupy a significant portion of mainstream media cov-
erage on Influencers. However, controversy-seeking Influencers who engage
in shame practices are a vocal, high-profile minority and not representative of
the larger segment of the industry—Influencers who generally aim to put their
best behavior forward and maintain their role-model status among followers.
Demonstrating some reflexivity on the permanence of information on the
web, a handful of Influencers also mused about reactions from their parents,
prospective employers, and future children if the top few search results for
their names raked in controversial material as a result of their self-surveillance,
or what Humphreys (2013) described as recording oneself for archival or
sharing purposes. However, these concerns were often merely passing com-
ments. When asked if they worried about the privacy of their personal infor-
mation archived on the web, few Influencers displayed any concern as they,
like Humphreysʼ informants, “implicitly defined privacy as privacy from other
users or people and not privacy from state, corporate or bureaucratic entities”
(2013, p. 6). In fact, many Influencers seem to have faith that new eruptions
of pseudo-events (Boorstin, 1961) and the cyclical spectacles that take the
form of social dramas (Turner, 1974) would quickly surpass their Influencer
wars and self-shaming, making them yesterdayʼs news.
Web Amnesia
Controversy-seeking Influencers may not always publicly discuss their con-
cerns about negative publicity. On the contrary, many invite it to capital-
ize on the attention. In my interviews, however, other Influencers perceived
L8r H8r 45
Influencer wars and negative attention rituals as effective but harming atten-
tion strategies. Many agree that “it is very important to stay relevant,” that
they “want to remain talked about,” and that they want to “differentiate”
themselves from others. Yet, they also value the ability to dissociate them-
selves from deviance over time. Although not always explicitly expressed,
many Influencers reference a sentiment of “forgetting” or what I term “web
amnesia:”
… the news changes so fast, it wonʼt even be relevant in a few days
… the [negative attention event] used to be the hottest news … we [would]
check forums and Tweets everyday, but I think not a lot of people remember
it now …
Unlike scholarly discussions that describe the infrastructure and technology of

the Internet as one that never forgets (Rosen, 2011) in light of data retention
tendencies, web amnesia is focused on the social effects followers experience
in the age of abundant data (Goldhaber, 1997). I posit here three vernacular
understandings of web amnesia that have emerged from my personal inter-
views and observations.
First, as observed by Goldhaber (1997), in the attention economy, there
is always an abundance of content that is rapidly circulating. This has been
exacerbated in recent years by increasing volumes of content produced via
new social media and messaging platforms not covered in this chapter (e.g.,
Snapchat, WeChat, LINE, and QQ). With spectacles and trends experiencing
a high turnover rate, even dramatic news gets old very quickly and loses its
impact on followers, resulting in lack of capacity to wrestle attention.
Second, there are typically several attempts at soliciting publicity in any
given period of time. As the Influencer industry in Singapore rapidly expanded,
some Influencers took to shaming practices as an attention-grabbing strategy
to distinguish themselves from others (Goldhaber, 1997). Multiple Influencer
wars and negative attention rituals often collide and appeal to different seg-
ments of Internet users. As such, whether oneʼs incident or shame practice
trends and receives the spotlight may be a matter of how controversial it is,
timing, or just plain luck. Only a selected few Influencers ever get propelled
into a national—or region-wide limelight.
Lastly, with the practice of Influencer wars and self-shaming becom-
ing popular and even blasé to desensitized followers, controversy-seeking
Influencers are pioneering new practices of clickbait (Blom & Hansen, 2015)
(e.g., staged leaked sex tapes or grotesque visuals from plastic surgeries) in a
bid for attention. In other words, the moral boundaries of shaming are ever
shifting. In 2010, it was largely taboo for Influencers to admit to having
46 Crystal Abidin
undergone plastic surgery. By 2012, such surgeries became renarrativized as

an intimate journey of sharing, commoditized for sponsorship, and normal-
ized among followers. As boundaries of what constitutes a spectacle keep
shifting, newer shaming practices will reinvent narrative scripts and innovate
to further solicit reactions and command attention.
L8r H8r
Some Influencers vie for attention through negative attention strategies
such as Influencer wars and indulge in inviting hating from followers. Yet,
as observed in my discussion of web amnesia, most Influencers have to man-
age a high rate of ephemerality in the spectacles they stage. For this reason,
Influencers deliberately strategize and labor over feedback loops comprising
their spectacle, reactions from followers, and responses to the hating they
receive in a self-fulfilling prophecy that continually generates new attention;
this is evidenced through taking and circulating screenshots of already-de-
leted faux pas, archiving and publicizing even the bad press they receive, and
provoking haters. In an environment where attention is scarce and increas-
ingly dispersed, Influencers rely on followers and haters, and on controver-
sy-seeking Influencers and each other to sustain an ecology of attention in
which moral boundaries are continually reasserted in order to be breached
through weaponized, vernacular, and reflexive shame, such that pseudo-events
(Boorstin, 1961) and social dramas (Turner, 1974)—or unsocial pseudodra-
mas—can continue to be produced as spectacles. As one veteran Influencer
told me, “all publicity is good publicity, even bad publicity … yeah only if you
know how to manage it.”
References
Abidin, C. (2013). Cyber-BFFs: Assessing women’s ‘perceived interconnectedness’ in
Singapore’s commercial lifestyle blog industry. Global Media Journal Australian
Edition, 7(1).
Abidin, C. (2015). Communicative <3 intimacies: Influencers and perceived interconnect-
edness. Ada: A Journal of Gender, New Media, & Technology, 8.
Abidin, C. (2017). Sex bait: Sex talk on commercial blogs as informal sexuality education.
In L. Allen & M. L. Rasmussen, (Eds.), Palgrave handbook of sexuality education
(pp. 493–508). London, UK: Palgrave Macmillan.
Alberoni, F. (2007). The powerless ʻeliteʼ: Theory and sociological research on the phe-
nomenon of the stars. In S. Redmond & S. Holmes (Eds.), Stardom and celebrity: A
reader (pp. 65–77). London, UK: Sage.
L8r H8r 47
Bird, E. (2003). The audience in everyday life: Living in a media world. London,
UK: Psychology Press.
Blom, J. N., & Hansen, K. R. (2015). Click bait: Forward—Reference as lure in online
news headlines. Journal of Pragmatics, 76, 87–100.
Boorstin, D. J. (1961). The image: A guide to pseudo-events in America. New York,
NY: Harper Colophon Books.
Burgess, J., & Green, J. (2009). YouTube. Cambridge, UK: Polity Press.
Daniel, E. V., & Knudsen, J. C. (1995). Mistrusting refugees. Berkeley: University of
California Press.
Davenport, T. H., & Beck, J. C. (2001). The attention economy: Understanding the new
currency of business. Boston, MA: Harvard Business School Press.
Debord, G. (2002). The society of the spectacle. Canberra, Australia: Treason Press.
Geraghty, C. (2007). Re-examining stardom: Questions of texts, bodies and performance.
In S. Redmond & S. Holmes (Eds.), Stardom and celebrity: A reader (pp. 98–110).
Los Angeles, CA: Sage.
Goldhaber, M. H. (1997). The attention economy and the net. First Monday, 2(4).
Retrieved from http://firstmonday.org/article/view/519/440
Hardacker, C. (2010). Trolling in asynchronous computer-mediated communica-
tion: From user discussions to academic definitions. Journal of Politeness Research,
6(2). 215–242.
Humphreys, L. (2013). Mobile social networks and surveillance: Usersʼ perspectives. In A.
Jansson & M. Christensen (Eds.), Media, surveillance, and identity: A social perspec-
tive (pp. 109–126). New York, NY: Peter Lang.
Kitzmann, A. (2004). Saved from oblivion: Documenting the daily form diaries to web cams.
MacAloon, J. J. (1984). Rite, drama, festival, spectacle: Rehearsals toward a theory of cul-
tural performance. Philadelphia: University of Pennsylvania Press.
Malefyt, T. de Waal, & Morais, R. J. (2012). Advertising and anthropology: Ethnographic
practice and cultural perspectives. London, UK: Berg.
Marshall, P. D. (2010). The promotion and presentation of the self: Celebrity as marker of
presentational media. Celebrity Studies, 1(1), 35–48.
Marwick, A. E. (2013). Status update: Celebrity, publicity, & branding in the social media
Papacharissi, Z.(2010). Privacy as a luxury commodity. First Monday, 15(8).
Phillips, W. (2015). This is why we canʼt have nice things: Mapping the relationship between
online trolling and mainstream culture. Cambridge, MA: MIT Press.
Rosen, J. (2011). Free speech, privacy, and the web that never forgets. Journal on
Telecommunications and High Technology Law, 9, 345–356.
Senft, T. M. (2008). Camgirls: Celebrity & community in the age of social networks.
Turner, G. (2010). Ordinary people and the media: The demotic turn. Los Angeles, CA: Sage.
48 Crystal Abidin
Turner, G. (2014). Understanding celebrity. Los Angeles, CA: Sage.

Turner, V. (1974). Dramas, fields, and metaphors: Symbolic action in human society. Ithaca,
NY: Cornell University Press.
Warner, M. (2002). Publics and counter publics. Quarterly Journal of Speech, 88(4).
413–425.
4.
Production and Performance of White
Anti-Racism in Online Media
Michael Potts
Just before Christmas 2013, public relations executive Justine Sacco boarded
a plane for the long-haul flight from London to visit family in South Africa.
She tweeted: “Just going to Africa. Hope I don’t get AIDS. Just kidding,
I’m white!” With only 170 followers on Twitter and no replies to any of her
previous tweets, Sacco did not expect much response. However, the tweet
came to the attention of Sam Biddle, a columnist for Gawker’s Valleywag
online magazine, who published it with the scathingly sarcastic title “And
Now, a Funny Holiday Joke from IAC’s PR Boss.” From there it was picked
up by Buzzfeed and other online magazines and spread like wildfire on social
media where the response was hyperbolic. “Justine Sacco should get fired …
and get AIDS” was a typically vitriolic tweet, and others gleefully gloated over
the prospect of her losing her position before she was even aware of what was
happening. By the time Sacco’s flight landed eleven hours later, the hashtag
#Hasjustinelandedyet? was trending worldwide (Vingiano, 2013). Biddle’s
original brief post had more than 200,000 shares on various social media sites
and would generate two further columns, both widely shared (Biddle, 2013).
The following day, Sacco’s employer, IAC (owner of The Daily Beast, Vimeo
and other well-known internet brands), fired her. From successful corporate
executive to global online notoriety to unemployment had taken less than
24 hours.
Why did this tweet among the millions of others tweeted every day
become such a phenomenon? The fact that Sacco was incommunicado, and
therefore unaware of the drama doubtless added piquancy to the internet
outrage, but does not explain it. Jon Ronson, who interviewed Sacco, Biddle,
and others for a book on public shaming, put it down in large part to the
fact that the seeming callousness of the tweet gave people the opportunity
50 Michael Potts
to demonstrate their own empathy by condemning her seeming lack of it. “I

think a lot of people really wanted to show people that they were empathetic”
he observed, “and so in their desperate desire to show they’re empathetic
they did this incredibly un-empathetic thing” (Miller, 2015). Slate’s Jordan
Weismann reflected in an end-of-year retrospective that the nature of online
media news production relying on shares via social media encouraged pander-
ing to this kind of mass outrage and gleeful condemnation. “I didn’t become
a journalist to peddle indignation on Facebook” he wrote, “but it sells—the
page views don’t lie. And while Slate may be running an entire special pack-
age dissecting the ways our perpetual umbrage colors and sometimes poi-
sons discourse, we play a role in it” (Weissman, 2014). Both of these factors
were necessary conditions for Sacco’s tweet metastasizing into a life-changing
event, but they were not by themselves sufficient. The final and perhaps the
most crucial element was the fact that as a successful public relations execu-
tive Sacco should have known better than to tweet something that could be
construed as crass and racist. She had transgressed the manners and mores of
acceptable middle-class discourse and paid a heavy price for it.
In this essay, I want to use the example of Justine Sacco as a springboard
from which to analyze the ways White identity is presented and policed online
and how this relates to the production and consumption of online media.
Applying Shannon Sullivan’s (2014) idea of “racial etiquette,” I argue that
much putatively anti-racist or social justice content production and sharing
online is actually catering to and driven by intra-racial politics of class. Building
on the work of Emma Kowal (2015 and Sara Ahmed (2004), I develop the
idea of White anti-racism as a specific type of performative behavior that, par-
adoxically, works to police White identity and retain control of the dialogue
around racial inequalities even as it purports to tackle them. Further, I argue
that a political economy underpins the click-driven advertising economy of
online media and that, ironically, much online news production purporting
outrage at racist faux-pas such as Sacco’s is actually about policing White
identity and behavior.
In Good White People: The Problem with Middle-Class White Anti-Racism
(2014) Shannon Sullivan argued that racial etiquette operates not only
inter-racially to govern interaction between different ethnicities but also
intra-racially as a form of defining, determining, and policing to what class
someone belongs. At the bottom of the hierarchy are what Sullivan called
“white trash,” those poor, under-educated Whites whose perceived coarse-
ness, ignorance, and slovenliness constitute a “breach of white social eti-
quette [that] threatens the boundary between white and nonwhite (especially
black) people” (2014, p. 30). White identity, then, is transposed onto class
The Performance of White Anti-Racism in Online Media 51
identity, with poor Whites seen as blurring or threatening ethnic bound-

aries. Conversely, as Sullivan noted, successful African Americans are seen
as confirming and validating the class-ethnicity transposition; it affirms lib-
eral Whites’ belief that “the basic structure of liberal society is just and fair
[whereas] the existence of white trash threatens both white liberal ideals of
opportunity and white liberal assumptions of openness to and acceptance of
people who are different than oneself” (p. 41).
Similarly, sociologist Kirby Moss found that many middle-class, oth-
erwise progressive, White liberals tended to use class or status as a marker
for now-unacceptable racial distinctions. After being told by an interviewee
named Carol that she didn’t notice racial or class distinctions but simply liked
to associate with successful people “like you,” Moss remarked,
My encounters with many middle-class participants in this research revealed the
way my privilege coupled with my Blackness served the unintended purpose of
empowering the very middle-class Whites I met … In their contact with me
[they] expanded their own representation of themselves (in particular, self-de-
scribed liberals or radicals like Carol) as being open and accepting. They … were
the ideal and I was the product of their ideals, a product … of the meta-ideology
of productivity and success … (Moss, 2003, p. 77).
The meta-ideology of middle-class White liberals such as Carol meant that

they had to reconcile their own privileged position with an acknowledgment
that inequalities persisted in society. Therefore, there was a tendency to iden-
tify racism and prejudice with poor, unhappy Whites (White trash) who were
bitter at their lack of success, and, conversely and consequently, to identify
their own success with open-mindedness and a willingness to accept people
on their own terms. To discriminate against someone because of their eth-
nicity or race would therefore be a gross breach of racial etiquette because it
would invalidate the perception that “liberal society is just and fair” (Sullivan,
2014, p. 41). It would expose the inequalities and privileges that underpin
the distribution of wealth and position in society.
Considered in this light, the rage against Justine Sacco’s tweet can be bet-
ter comprehended. Sacco came from a prominent, progressive, middle-class
White South African family that supported Nelson Mandela’s African National
Congress party. A senior director of communications at only 30 years of age,
Sacco was the epitome of successful White middle-class America. Her priv-
ilege demanded that she observe the rules of her class, especially racial eti-
quette, seen as a particularly sensitive area due to ongoing racial inequalities.
Ironically, Sacco was in fact attempting to signal her awareness of these
issues by parodying the obliviousness people like her often had to the harsh
realities of life in other parts of the world—a subtlety which did not translate
52 Michael Potts
well to the 140 characters of Twitter. Instead, read by those who did not
know her, it was perceived and sometimes actively misconstrued as embrac-
ing White privilege and mocking inequalities (Ronson, 2015). Such a breach
of White social etiquette would threaten to undermine the perception that
those in privileged positions in society were cognizant of their privilege and
working for the greater good by breaking down residual barriers to equality.
Sacco’s tweet was presented as making a mockery of that unspoken etiquette,
and the chastisement was astonishingly swift and harsh. Just like all public
shaming, it worked to police behavior by encouraging mutual and vigorous
public condemnation. By sharing the story people were able to demonstrate
their own decency by denouncing both the transgression and the transgres-
sor; being summarily fired and fleeing into self-imposed exile (Sacco took a
temporary position as an NGO worker in a remote Ethiopian village without
electricity) was an effective public punishment.
For all of the virtual column inches generated by and responding to the
tweet, what has been left largely unanalyzed is the extent to which the furor
was created, fed, and sustained by online media personalities and brands.
They brought what was a tweet for a few followers to global attention and
then collated, re-published, and reported on the responses. Discussing the
incident with Ronson, Salon writer Judith Miller (2015) addressed the prob-
lem of blurring the line between journalists as putatively objective profes-
sional and journalists as career-focused individuals concerned with building
their own personal brand in part by tweeting their own opinions. Media per-
sonalities might comment on social media as though expressing their own
private opinions, but these opinions are in fact carefully filtered and curated.
The scandal was, to a significant extent, manufactured by those commenting
and responding to it.
This model of essentially manufacturing outrage by presenting an incident
or remark out of context, encouraging condemnation and then reporting on
the condemnation in a cycle of outrage on outrage, had been used elsewhere
to appeal to a similar audience of college-educated, politically progressive,
middle-class Whites. Even as the Sacco controversy was generating millions
of page-views for Vice and other online media, the relatively staid political
commentary site PolicyMic was discovering that stories about race and racism
generated a powerful emotional response which translated to a substantial
increase in page-views and shares on social media. The site changed its name
to the shorter, hipper Mic, abandoned its initial studied objectivity, and began
writing stories that monetized concerns about social justice:
In its early days the site published left-leaning stories alongside right-leaning
takes like “Is There a Media Bias Against Ron Paul” … It also trafficked in
standard content farm fare like “The 25 Greatest Things About Christmas.” The
site also started to develop an inertia around a certain type of story: simple, emo-
tional social justice narratives … These stories got traction on Facebook, so Mic
replicated them, attracting more social justice readers as well as more social jus-
tice writers, who then wrote more social justice stories. “Mic realized earlier than
most places that they could commodify people’s feelings about race and gender”
was the view of one early staffer. (Jeffries, 2017)
As a former Mic employee related, the aim of such pieces was not to analyze,
militate for action, or even raise awareness as such, but rather to generate
shares on social media by sparking a visceral emotional reaction and giving
people the opportunity to display their own values by sharing and condemn-
ing an example of reprehensible behavior:
“Mic trafficked in outrage culture,” a former staffer who left in 2017 said. “A
lot of the videos that we would publish would be like, ‘Here is this racist per-
son doing a racist thing in this nondescript southern city somewhere.’ There
wouldn’t be any reporting or story around it, just, ‘Look at this person being
racist, wow what a terrible racist.’ ” (Jeffries, 2017)
Alongside outrage, Mic actively sought out stories presenting Black Americans
in such a way as to confirm the desire of White, middle-class readers to see
their values reflected, and to have middle-class society confirmed and vali-
dated as essentially fair and meritocratic. In one instance, a reporter pitching
a story about a woman who was building rooftop gardens was eagerly inter-
rupted by then editor-in-chief Jake Horowitz demanding, “But is she black?
Is she black?” (Jeffries, 2017 emphasis in original). Horowitz had a keen eye
for the angle that would manifest and confirm his readers’ views and ensure
that the story would be widely shared.
The formula for success used by Mic, Vice, and numerous other online
media companies relied on the fact that people share stories that resonate
emotionally and allow them to display their values and ideology. Lee, Ma, and
Goh (2011), for example, found that status enhancement is a primary moti-
vation in selecting and sharing news stories. Furthermore, they found that
involvement in social media also reshaped users’ interaction with news media,
driving an increase in performative behavior where certain news stories were
shared so as to present the person sharing in a desirable light.
This finding builds on research demonstrating that social media users felt
significant social pressure to curate their online presence to represent them-
selves in ways that would advance their social standing. For instance, Park,
Kee, and Valenzuela found that college students “joined [Facebook] groups
because they felt peer pressure, wanted to make themselves look cool, and to
develop their career” (2009, p.731).
54 Michael Potts
Online media, therefore, can be said to fulfill at least two significant func-
tions. The primary or ostensive function is to disseminate news that the user
deems relevant or interesting, and a secondary but nevertheless important
function is to portray the user as a person who cares about the issues outlined
in the story. Digital media outlets such as Mic can and do tailor their stories to
maximize shareability and hence page-views and advertising revenue. Social
media users tend to be younger, better educated, more liberal-progressive in
outlook, and more politically engaged than the population at large (Mellon
& Prosser, 2017). Kruse, Norris and Flinchum (2018) found social media
users post content they believe others will agree with in order to avoid dis-
agreement and advance social standing. This tendency can lead to a situation
known as an echo chamber or hug box where posting content becomes more
about signaling conformity than sharing interesting, relevant, or intellectually
stimulating or challenging content.
All in all, therefore, it seems news stories are often not only shared on
social media to signal political ideology and advance status but also written
with the aim of satisfying this tendency.
Crowdsourcing Bigotry: Manufacturing Controversy in

Digital Media
As is perhaps to be expected, a fundamental part of signaling political stance
and advancing in-group behavior is identifying and anathematizing a com-
mon enemy. Social media has greatly facilitated this process, allowing a con-
troversial opinion on almost anything to be sourced with a simple search.
Marketers and corporations have recognized the potential of a controversy
which, if it takes off, can generate free coverage worth hundreds of thousands
of dollars. Alex Nichols (2018) observed that elevating the rumblings of a
few malcontents on social media sites such as Reddit can generate viral stories
that resonate with readers by provoking outrage at the seeming prevalence
of bigotry in society at large. A handful of bigoted remarks can be quickly
sourced and aggregated into a story that generates a visceral response from
readers who are likely to share it:
Manufactured outrage campaigns of this sort aren’t new […] Superficial messages
of women’s empowerment are relatively apolitical, having been co-opted long ago
by dubious figures like Sheryl Sandberg and Ivanka Trump. Taking a stand on
long-settled issues like “women should be allowed in TV commercials” brings out
a relatively small opposition of revanchist losers and opportunistic trolls, whose
unsympathetic whining lends itself to aggregation in “Men are mad about X”
trend pieces. This amplification of fringe voices imbues the mundane act of giving
money to KFC or Warner Bros. with political meaning. (Nichols, 2018)
Digital media companies have used this strategy to generate stories that eschew
nuance and a thorough interrogation of the structural nature of inequality in
favor of simple, emotive social justice narratives that scour social media sites
for quotes that reduce the problem to poorly educated, bad Whites blocking
the path to progress and equality.
For example, reviewing online media’s response to Seattle Seahawks’
Richard Sherman’s intemperate end-of-game behavior, Abraham Iqbal Khan
(2016) noted that articles from outlets such as Deadspin and The Nation
responded fiercely to a racist backlash that was almost entirely imaginary.
Following a bad-tempered game between the 49ers and the Seattle Seahawks,
Sherman had extended a hand to the 49ers’ Michael Crabtree, only to be
rebuffed with a contemptuous push to the face. Interviewed soon after,
Sherman was clearly angered by the disrespect, and in a WWF-style throw-
down challenge addressed the camera directly, shouting “I’m the best corner
in the game … don’t you ever talk about me” (Brinson, 2014). It was a
momentary outburst that he later ascribed to previous bad feelings between
the players (Sherman, 2014).
Feelings ran high off-field as well, with fans on both sides attacking the
behavior of the opposing player. At times, this descended into outright abuse
on social media that took on an ugly racial tone. This gave digital media
outlets the opportunity to piggyback on a trending topic with a controversy
that was effectively manufactured for public consumption. Deadspin’s arti-
cle “Dumb People Say Stupid, Racist Shit about Richard Sherman” handily
supplied readers with the requisite reaction to the content, which consisted
entirely of derogatory and bigoted tweets from otherwise unknown users.
As with the Mic videos of otherwise unknown people being racist without
any context, analysis, or purpose, Deadspin’s article was no more than a col-
lation of tweets for the sake of it, with only seven sentences of text providing
the barest contextualization together with the obligatory condemnation to
make clear where the writer and the subsequent sharers’ proper sympathies
lay. As Khan noted, “a handful of hateful tweets” gave online media “the
opportunity to exhibit their anti-racist credentials in torrents of self-refer-
ential speech” (2016, p. 41). In short, the story required racist reaction to
Sherman’s outburst, so racist reaction was sourced from hundreds of millions
of tweets to supply the material for a story on the prevalence of racism on
social media that relied for its virality on the energy and outrage it generated.
Such articles have virality because they generate a visceral sense of out-
rage and rely on social media’s tendency to facilitate the formation of “ste-
reotyped, value-laden judgements about the political out-group” that “run
the gamut from intelligence to skill level to moral values” (Settle, 2018,
56 Michael Potts
p. 197). Stories such as this allow strong in-group signaling by supplying

a caricature of an out-group that reveals it to be monstrous, close-minded,
and callous. The incident itself is often trifling. Its importance stems not
from the event but from the fact that it performs a kind of positional work
that allows sharers to demonstrate not only that they possess the proper
sympathies and attributes but also that their group is meritocratic: the head-
line proclaimed that it was not just racist people who said racist things about
Richard Sherman, but “dumb people” who said “stupid, racist things”
(Kalaf, 2014).
Thus, despite the initial difference from the Sacco story, the Sherman
incident strongly suggests the same inference should be drawn: that there
is a political economy at work here which underpins the actual economy of
sharing on social media, page-views, and advertising revenue on which dig-
ital media rely. As I argue below, this is a form of status signaling and racial
etiquette that falls under the rubric of White anti-racism, itself an attenuated
form of middle-class racism that polices so-called proper White, middle-class
behavior.
Policing White Identity: The Paradoxical Nature of White

Anti-Racism
In a recent New Yorker article, Katy Waldman reviewed Robin DiAngelo’s
White Fragility and considered DiAngelo’s surprising and somewhat confron-
tational contestation that it is middle-class White liberals who are most cul-
pable for perpetuating racial inequality by placing themselves beyond or past
race and ethnicity, and subsequently locating all the blame for racial injus-
tice at the doorstep of overtly bigoted racist Whites. Waldman recounted
DiAngelo’s thesis that White fragility (defensiveness against imputation of
racism) correlates with the public performance of anti-racism as part of a
specific identity:
[DiAngelo] reserves her harshest criticism for white liberals … whom she sees
as refusing to acknowledge their own participation in racist systems. “I believe,”
she writes, “that white progressives cause the most daily damage to people of color.”
Not only do these people fail to see their complicity, but they take a self-serving
approach to ongoing anti-racism efforts: “To the degree that white progressives
think we have arrived, we will put our energy into making sure that others see
us as having arrived” … DiAngelo writes [that] white people cling to the notion
of racial innocence, a form of weaponized denial that positions black people as
the “havers” of race and the guardians of racial knowledge. Whiteness, on the
other hand, scans as invisible, default, a form of racelessness. (Waldman, 2018,
emphasis in original)
DiAngelo identified a tendency on the part of many White liberals to assign

race to others but not to themselves. Although perhaps well intentioned, it is
a stance that presumes to transcend not only race but also ethnicity and even
nationality. It sets up implicit hierarchy of care and claims to value in others
what it disclaims for itself. Most saliently, the tendency functions as a marker
for a set of political and cultural attributes: White, educated, middle class, and
self-consciously progressive.
Examining this tendency among non-governmental organizations and
charity and government workers working with indigenous Australians,
anthropologist Emma Kowal analyzed how one ritualized anti-racist speech-
act—performing or participating in a “welcome to country”1—serves to rein-
force an attenuated sense of belonging and identity that works by disavowing
the very function it performs: such acts, she argued, “maintain White iden-
tities and manage White stigma by questioning White belonging” (Kowal,
2015, p. 180). In other words, such speech-acts work by signaling belonging
within a group of people who “paradoxically experience belonging through a
sense of not belonging” (p.180). In their performance, we can discern both
the ostensive function of White anti-racism (in this case, signaling respect and
valuing of the ethnic other as ethnic other) and an unspoken function of rein-
forcing proper White identity along lines of class, culture, and ethnicity.
Unpacking this further, Sara Ahmed (2004) has written about White
anti-racism as a non-performative performative act (i.e., an act that does not
or cannot do what it claims to do). Ahmed’s avenue of inquiry is particu-
larly fruitful, because she opens up how the performance of White anti-rac-
ism operates from an implicit perspective of being beyond race and ethnicity,
thereby placing the person performing it in a putatively objective place to
assign race and ethnicity together with moral valence: to decide who counts
as having an acceptable racial identity, which expressions of racial or cultural
identity are permissible, and so on. It is a “fantasy of transcendence” in which
de-racialized Whiteness “scans as invisible, default, a form of racelessness”
(Ahmed, 2004). Standing outside of race and ethnicity has powerful benefits
and fundamentally changes the directionality of anti-racism, placing the per-
former in the position of critiquing rather than being critiqued. Hence, even
problematic behavior on the part of a person or institution can become part
of the performance of White anti-racism.
To illustrate this point, Ahmed described her own experience writing her
university’s race equality policy. She found that the very performance of the
act allowed the university to step out from the umbra of being a racist insti-
tution and simultaneously become morally competent to enforce or prohibit
racial and cultural identity, to demarcate what was appropriate and assign it
58 Michael Potts
moral value. Despite the university’s own relatively tenuous moral standing
(as is the case of most Western universities, historically speaking), it becomes
the arbiter of racial identity rather than the perpetrator of racism.
Ahmed shrewdly observed that what happens in such situations is a form
of displacement, a shifting of focus in which the person or institution under-
taking the performance of White anti-racism can re-direct the gaze from
themselves to others, or, at least, obviate blame by becoming putatively part
of the solution. Throughout this process, critical ethical and class distinctions
are being made:
antiracism becomes a matter of generating a positive white identity, an identity
that makes the white subject feel good about itself. The declaration of such an
identity is not in my view an anti-racist action. Indeed, it sustains the narcissism
of whiteness and allows whiteness studies to make white subjects feel good about
themselves, by feeling good about “their” antiracism. One wonders again what
happens to bad feeling in this performance of good, happy whiteness … I suspect
that bad feelings of racism (hatred, fear, pain) are projected onto the bodies of
unhappy racist whites, which allows progressive whites to be happy with them-
selves in the face of continued racism towards non-white others. (Ahmed, 2004)
Ahmed argued that White anti-racism is thus frequently non-performa-

tive in the sense of directly combating racism, but that it is performative in
that it displaces culpability and complicity in racism and racist society. The
work it actually performs is not its ostensive work but its “other” work. White
anti-racist speech-acts, she claimed, “have a very specific form: they define
racism in a particular way, and then imply ‘I am not’ or ‘we are not’ that”
(Ahmed, 2004). Similarly, Kowal’s analysis of a welcome to country as a ritual-
istic White anti-racist speech-act shows that its performance, although super-
ficially about the ethnic other, is also about drawing distinctions between the
good anti-racist Whites participating and the bad Whites who perpetrate and
perpetuate racism elsewhere.
White anti-racism, then, facilitates the containment of dissonance. It
is not merely a case of pretending to be what one is not but of compart-
mentalizing concern, of being outraged in the abstract about prejudice and
social injustice but partaking in and benefiting from exclusionary practices in
specific instances. It can be seen as analogous to the performance of White
Christian moral rectitude in America, and the obscene (in a Lacanian sense)
transgressions licensed by the performance of moral probity. It is hypocriti-
cal, of course, but it is not simply hypocrisy. Psychologically and sociologi-
cally, it is more complex than that. It is the rationalization of privilege and
subtle discrimination by ostentatiously taking up arms against the cruder,
more obvious and (crucially) the pettier forms of privilege and discrimination
perpetrated by “poor White trash.” It is not a lesser form of racism, but a less
obtrusive and more managerial way of handling racial hierarchy that allays the
concomitant psychological and emotional turbulence.
Social media and sharing digital media stories as a demonstration of proper
sympathies greatly facilitate and streamline this process. Stories, op-eds, and
articles that claim to expose, document, or tackle racism allow the person
sharing them to both share in the sense of moral indignation that such things
still occur and document their own sensibilities. In Ahmed’s formulation,
sharing such a story allows the sharer to implicitly define racism as that which
is exhibited in the material being shared and by condemning it imply: “I am
not this.” Because digital-only media outlets often rely solely on advertis-
ing revenue and look to social media to generate page-views, they may be
tempted to frame news stories with shareability in mind. News becomes a
cultural product as much as an information product, allowing social media
users quickly and effortlessly to express something about themselves (they are
caring and politically aware) that is otherwise difficult and time-consuming
to express.
Alongside this temptation is a parallel tendency for journalists to market
themselves in quasi-cultural ways, rather than solely as objective and reliable
observers, investigators, and reporters. This process is greatly facilitated by
social media, as Slate’s Andrew Goldman reflected:
Before I joined it, and before it ruined my life, I hated Twitter. It struck me as a
baldly narcissistic waste of time … But in 2011, after I got the most high profile
gig of my life—doing the weekly Q&A in the New York Times Magazine—I
decided that I would now need to become a “brand” unless I wanted to be pas-
tured after my 40th birthday like so many aging media types before me. As soon
as I joined, I became addicted. I rationalized away my previous hatred, because,
like pretty much everyone else I know who makes a living with a byline, I’m a
desperate narcissist too. (Goldman, 2014.)
Social media allow journalists and writers, along with everyone else, direct
and unmediated access to the public forum. It also allows them to build a fol-
lowing and an identity as a distinct voice or brand. Although reporters and
commentators have had particular interests and perspectives for as long as
there has been news, social media, together with the increasing casualization
and turnover of media workers, have made it imperative that reporters build
their own brand outside of the company for which they work. Social media
gives them visibility, keeps them connected to the wider network of media
producers and consumers, and serves as a kind of informal publicly accessible
portfolio of work. It also means though that there is strong pressure to filter
and package what they say with a view to consumption.
60 Michael Potts
Together, these tendencies mean that increasingly news stories are shaped,
wittingly or not, by their cultural utility instead of, or as well as, their news
value. Stories with little inherent news value can still be heavily shared on
social media because of their positional value: they allow readers to posi-
tion themselves culturally and ideologically. Alongside this, there has been an
increase in stories that begin on social media and largely play out there, but
nevertheless make the headlines of national newspapers and respected digital
media outlets. Because such stories have their genesis on social media, the
continuing reporting of such incidents can be heavily inflected by the initial
responses, particularly from journalists, commentators, and other influencers
who are cognizant not only of how the story will be perceived but also how
any comment they make on it will be perceived.
A salient example of this trend occurred in 2019, when a group of boys
from a Catholic school went to a pro-life protest march in Washington DC.
Video posted to Twitter appeared to show the students gathered on the steps
of the Lincoln Memorial, with one of them blocking the path of Nathan
Philips, a Native American man, while others gathered around in a mocking
and disrespectful manner. Some of the schoolboys wore Make America Great
Again (MAGA) hats, seemingly confirming this as a case of entitled hooligan-
ism and bullying of a minority person—a suspicion soon confirmed by Philips
who claimed he had been trying to calm a fractious situation and that the
students had surrounded and threatened him.
Initial response from well-known public figures on Twitter was splenetic.
New York Times columnist Kara Swisher saw toxic masculinity in the “fetid
smirking harassing” and invited those who saw it differently to “go fuck your-
selves” (Swisher, 2019). Vulture writer Erik Abriss tweeted “I just want these
people [the schoolboys] to die. Simple as that. Every single one of them.
And their parents” (Levine, 2019). Film producer Jack Morrisey tweeted that
the “kids [should go] screaming, hats first into the woodchipper,” though
he later apologized calling it “just a fast, profoundly stupid tweet” (Hod &
Levine, 2019). Comedian Kathy Griffin posted a photo of Covington school-
children holding up three fingers at a basketball game (for a three-point shot)
and feverishly accused them of “throwing up the new Nazi sign,” though she
later deleted the tweet (Ernst, 2019).
The intensity of the reaction from journalists and other commentators on
social media ensured that the story was covered not as a minor incident at a
protest, but as an indictment of racist American society, with bigoted White
hooligans descending on and mocking a hapless older indigenous protestor.
However, more footage soon emerged, which contextualized the seeming
confrontation between the schoolboy and the Native American protestor,
showing initial coverage to have been unnuanced at best and disingenuous at

worst. An editor’s note appended to an article in The Washington Post noted
that “certain statements reported by Phillips are not corroborated by widely
circulated video of the incident” (Miller, 2019). A further, separate editorial
note published in the paper accepted that,
Subsequent reporting, a student’s statement and additional video allow for a
more complete assessment of what occurred, either contradicting or failing to
confirm accounts provided in that story—including that Native American activist
Nathan Phillips was prevented by one student from moving on, that his group
had been taunted by the students in the lead-up to the encounter, and that the
students were trying to instigate a conflict (“Editor’s Note”, 2019).
It seemed Philips had wandered into the middle of the school group
waiting for the bus. The boys, although excitable and immature, had not tried
to stop or harass him but rather responded with a mixture of bemusement
and carnivalesque license. But the video showing MAGA hat-wearing White
schoolboys apparently going out of their way to mock and bully an older
Native American man was a compelling opportunity for commentary and
condemnation.
As Caitlin Flanagan (2019) shrewdly observed, from the “patronizing
gentleness of the news media” toward Philips to the “celebrities tweet[ing]
furiously, desperate to insert themselves into the situation in a flattering
light,” the story was never about what was at most trifling stand-off; it was
always about the reaction to the story and the opportunity for demonstrating
proper progressive sympathies. For many people, Philips licensed a dated,
simplistic fantasy of the wise Native American elder reproaching a materialistic
and aggressive society. “Why is the Covington Catholic controversy still the
nation’s biggest story?” a Vox editorial asked, finding that the various videos
of the event became a kind of Rorschach test for political ideology and as such
gained crucial importance for culturally and ideologically positioning oneself
via social media (Beauchamp, 2019). The story had become a shibboleth, in
which expressing condemnation of what it seemed to represent was vastly
more important than the incident itself. It was a password to gaining accep-
tance in social groups that mattered in terms of professional advancement.
Similarly, a 2018 story about journalist Sarah Jeong’s appointment to
the New York Times served as a vehicle for the display of values as a proxy for
social standing. Old tweets of Jeong’s from her time at The Verge surfaced,
reflecting a series of inflammatory remarks about White people. “White peo-
ple have stopped breeding. You’ll all go extinct soon,” “#cancelwhitepeo-
ple,” and “dumbass fucking white people marking up the internet with their
opinions like dogs pissing on fire hydrants” were typical of what Andrew
62 Michael Potts
Sullivan (2018) of New York Magazine called Jeong’s “eliminationist rheto-

ric.” Inevitably, there was an outcry from people who felt that in appointing
someone with a history of denigrating a racial group the New York Times was
employing double standards. The incident was even more of a shibboleth
than the confrontation between the schoolboys and the Native American pro-
testor, because it became a mark of sophistication to indicate understanding
that Jeong’s remarks were anti-racist. Zack Beauchamp’s defense of Jeong’s
tweets in Vox exemplified this process and illustrates how the performance
of White anti-racism polices White identity at the same time as it claims to
question and challenge it.
Beauchamp (2018) argued that people who took offense did so because
they were taking the tweets literally and failed to understand that they were
meant “satirically and hyperbolically to emphasize … the ways in which a
power structure that favors white people continues to exist,” Jeong’s ref-
erences were, he argued, “shorthand,” “quasi-satirical generalizations” that
had to be “interpret[ed] … through social context” and were “obvious to
people who have been listening to these social media conversations.” Thus,
when she tweeted about “dumbass fucking white people” posting on social
media “like dogs pissing on fire hydrants” she was, properly understood,
“commenting on the ubiquity of (often uniformed) white opinion on social
media” (Beauchamp, 2018). In other words, people who had taken offense
were uninformed, not knowing how to interpret quasi-satirical generaliza-
tions through the prism of social context.
Beauchamp’s explication and defense of Jeong’s tweets demonstrates
how White anti-racism generates implicit distinctions based on sophistica-
tion, subsequently implying an intellectual and meritocratic division between
those who get it and those who do not. It sets up something of a dou-
ble-bind: if you accept that tweets such as #cancelwhitepeople are actually
anti-racist it shows that you are sophisticated enough to parse statements in
terms of discourse, to weigh social power dynamics, and to consider inflec-
tions of meaning and rhetoric. If you demur at them, then by definition you
do not understand them (or are actively and defiantly taking them in delib-
erate bad faith). Intellectual sophistication is yoked to the performance of
moral probity, neatly invalidating other opinions as uninformed, bigoted, or
both. Distinctions of White racist and White anti-racist, therefore, become
questions of class and meritocracy and police the boundaries of acceptable
middle-class White opinion.
White anti-racism, then, can be seen as an archetypal form of post-mod-
ern discourse because it is entirely capable of functioning at a meta or ironic
level. As the Columbia Journalism Review pointed out, beyond the politically
partisan reactions to the appointment, it was well understood by journalists

that Jeong’s personal success was predicated on her cultivation of an edgy
but carefully calibrated and curated brand which appealed to sophisticated
professionals who understood that what she was doing was part of an act
and not to be taken literally (Uberti, 2018). One didn’t have to, in this case,
demonstrate full-throated support of Jeong personally. It was enough to sig-
nal that you understood the logic of the performance and the implications of
accepting it or not.
Atlantic and National Review editor Reihan Salam observed that such
rhetoric, although it might ostensibly seem to denounce privilege, actually
serves as a powerful coded signal of privileged class status:
The people I’ve heard archly denounce whites have for the most part been
upwardly-mobile people who’ve proven pretty adept at navigating elite, predom-
inantly white spaces. A lot of them have been whites who pride themselves on
their diverse social circles and their enlightened views, and who indulge in their
own half-ironic white-bashing to underscore that it is their achieved identity as
intelligent, worldly people that counts most, not their ascribed identity as being
of recognizably European descent. (Salam, 2018)
Once again, the myth of a meritocratic liberal society that Kirby Moss
observed as a crucial part of middle-class White identity is reinforced by
denouncing unearned White privilege, thus signaling that the position
enjoyed by the person expressing (or sharing) it is the result of intelligence,
hard work, and a cosmopolitan, open-minded attitude. Being able to produce
stories that allow people to quickly and effortlessly express one’s identifica-
tion with such values, to indicate that they are one of the people who gets it,
can have real value.
White anti-racism in the sense that Sullivan, Kowal and others use
it, and as it relates to social media, can be thought of as a post-modern
dynamic that purports to be concerned with racial and ethnic injustices,
but actually serves to inscribe complex signals of social status, hierarchy,
and ideology. Although it may instantiate a genuine urge to inveigh against
racism and prejudice, it tends to locate these phenomena elsewhere, thereby
redirecting attention and interrogation away from many of the actual loci
of power in society. Its influence on the production and consumption of
content on social media seem significant: I used the transition of Policy Mic
to Mic as an example, but there are many more. Salon, for instance, fol-
lowed a similar trajectory. As traditional media continue to suffer attrition
and news production and distribution continues to move online, we should
remember the cultural and social element to the production of online news
and opinion. Questions of how race, class, and ethnicity are reinforced and
64 Michael Potts
reinscribed by a political economy of sharing on social media that works by

claiming to challenge them cannot be ignored.
Note
1. In Australia, a “welcome to country” is a recognition and acknowledgment of prior
indigenous occupation and stewardship of the land before the arrival of Europeans.
References
Ahmed, S. (2004). Declarations of whiteness: The non-performativity of anti-racism.
Borderlands e-journal, 3(2). Retrieved from http://www.borderlands.net.au/index.
html
Beauchamp, Z. (2018). In defense of Sarah Jeong. Retrieved from https://www.
vox.com/policy-and-politics/2018/8/3/17648566/sarah-jeong-new-york-
times-twitter-andrew-sullivan
Beauchamp, Z. (2019). The real politics behind the Covington Catholic controversy, explained.
Retrieved from https://www.vox.com/policy-and-politics/2019/1/23/18192831/
covington-catholic-maga-hat-native-american-nathan-phillips
Biddle, S. (2013). And now, a funny holiday joke from IAC’s PR boss. Retrieved from http://val-
leywag.gawker.com/and-now-a-funny-holiday-joke-from-iacs-pr-boss-1487284969
Brinson, W. (2014). Richard Sherman calls Michael Crabtree ‘sorry receiver’ in wild rant.
Retrieved from https://www.cbssports.com/nfl/news/richard-sherman-calls-
michael-crabtree-sorry-receiver-in-wild-rant/
Editor’s Note related to Lincoln Memorial Incident. (2019, March 1) Washington
Post. Retrieved from https://www.washingtonpost.com/nation/2019/03/01/
editors-note-related-lincoln-memorial-incident/?utm_term=.10f8c355488a
Ernst, D. (2019). Kathy Griffin deletes ‘Nazi sign’ smear from stream of Covington Catholic
tweets. Retrieved from https://www.washingtontimes.com/news/2019/jan/22/
kathy-griffin-deletes-nazi-sign-smear-from-stream-/
Flanagan, C. (2019). The media botched the Covington Catholic story. Retrieved from https://
www.theatlantic.com/ideas/archive/2019/01/media-must-learn-covington-
catholic-story/581035/
Goldman, A. (2014). How outrage changed my life. Retrieved from http://www.slate.
com/articles/life/culturebox/2014/12/the_year_of_outrage_2014_everything_
you_were_angry_about_on_social_media.html#part-8
Hod, I., & Levine, J. (2019). Jack Morrisey apologizes for deleted Covington ‘woodchipper’ tweet.
Retrieved from https://www.thewrap.com/film-producer-jack-morrissey-apologizes-
for-deleted-covington-woodchipper-tweet/
Jeffries, A. (2017). Mic’s drop: How Mic.com exploited social media for clicks and then aban-
doned a staff that believed in it. Retrieved from https://theoutline.com/post/2156/
mic-com-and-the-cynicism-of-modern-media?zd=2&zi=bemuxzom
Kalaf, S. (2014). Dumb people say stupid, racist shit about Richard Sherman. Retrieved
from https://deadspin.com/dumb-people-say-stupid-racist-shit-about-richard-
sherm-1504843629
Khan, A. (2016). A rant good for business: Communicative capitalism and the capture of
anti-racist resistance. Popular Communication, 14(1), 39–48.
Kowal, E. (2015). Welcome to country: Acknowledgement, belonging and White anti-rac-
ism. Cultural Studies Review, 21(2), 173–204.
Kruse, L., Norris, D., & Flinchum, J. (2018). Social media as a public sphere? Politics on
social media. The Sociological Quarterly, 59(1), 62–84.
Lee, C., Ma, L., & Goh, D. (2011). Why do people share news in social media? In Zhong,
N., Callaghan, V., Ghorbani, A., & Hu, B. (Eds.), Active Media Technology: 7th
International Conference, AMT 2011, Lanzhou, China, September 7–9, 2011.
Proceedings (Vol. 6890, Lecture Notes in Computer Science, pp. 129–140). Berlin,
Heidelberg: Springer Berlin Heidelberg.
Levine, J. (2019). Vulture writer who wished death on Covington students fired from job at
INE Entertainment. Retrieved from https://www.thewrap.com/vulture-writer-on-
covington-students-i-just-want-these-people-to-die-simple-as-that/
Mellon, J., & Prosser, C. (2017). Twitter and Facebook are not representative of the
general population: Political attitudes and demographics of British social media users.
Research & Politics, 4(3).
Miller, J. (2015). Jon Ronson on social media witch hunts: “We’re the merciless ones!” retrieved
from https://www.salon.com/2015/03/29/jon_ronson_on_social_media_witch_
hunts_were_the_merciless_ones/
Miller, M. (2019, January 20). Viral standoff between a tribal elder and a high schooler is
more complicated than it first seemed. Washington Post. Retrieved from https://www.
washingtonpost.com/local/social-issues/picture-of-the-conflict-on-the-mall-comes-
into-clearer-focus/2019/01/20/c078f092-1ceb-11e9-9145-3f74070bbdb9_story.
html
Moss, K. (2003). The color of class. Philadelphia: University of Pennsylvania Press.
Nichols, A. (2018). Lady Doritos and the new facile culture wars. Retrieved from https://
theoutline.com/post/3328/lady-doritos-controversy-outrage-marketing?zd=
3&zi=wxe5zwd2
Park, N., Kee, K., & Valenzuela, S. (2009). Being immersed in social networking environ-
ment: Facebook groups, uses, gratifications, and social outcomes. CyberPsychology &
Behavior, 12(6), 729–733.
Ronson, J. (2015, February 15). Feed frenzy. The New York Times Magazine, pp. 20–28.
Salam, R. (2018). The utility of white-bashing. Retrieved from https://www.theatlantic.
com/ideas/archive/2018/08/the-utility-of-white-bashing/566846/
Settle, J. (2018). Frenemies: How social media polarizes America. Cambridge, UK,
Cambridge University Press.
Sherman, R. (2014). To those who would call me a thug or worse. Retrieved from https://
www.si.com/2014/01/20/richard-sherman-interview-michael-crabtree
66 Michael Potts
Sullivan, A. (2018). When racism is fit to print. Retrieved from https://nymag.

com/intelligencer/2018/08/sarah-jeong-new-york-times-anti-white-racism.
html?utm_source=tw&utm_medium=s3&utm_campaign=sharebutton-t&gtm=top
Sullivan, S. (2014). Good white people: The problem with middle-class white anti-racism.
Albany, NY: State University Press of New York.
Swisher, K. [karaswisher]. (2019, January 20). And to all you aggrieved folks
[tweet]. Retrieved June 14 2019 from https://twitter.com/karaswisher/
status/1086735766997344261
Vingiano, A. (2013). This is how a woman’s offensive tweet became the world’s top
story. Retrieved from https://www.buzzfeednews.com/article/alisonvingiano/
this-is-how-a-womans-offensive-tweet-became-the-worlds-top-s
Waldman, K. (2018). A Sociologist examines the “White fragility” that prevents White
Americans from confronting racism. Retrieved from https://www.newyorker.
com/books/page-turner/a-sociologist-examines-the-white-fragility-that-pre-
vents-white-americans-from-confronting-racism
Uberti, D. (2018). Sarah Jeong, The New York Times, and the Gamergate school of
Journalism. Retrieved from https://www.cjr.org/analysis/sarah-jeong-new-york-
times-gamergate-school-of-journalism.php
5.
Networked Gatekeeping
and Networked Framing of
#BlackLivesMatter Publics during
the 2016 US Presidential Election
Sharon Meraz
The Twitter hashtag #BlackLivesMatter was created in 2013 by Alicia Garza,

Patrisse Cullors and Opal Tometi in the wake of George Zimmerman’s
acquittal for the murder of an unarmed, young Black man, Trayvon Martin,
in Florida. The hashtag gained sustained fervor in 2014 after the killing of
Michael Brown in Ferguson, Missouri, when virtual anger spilled into face-to-
face protests on the streets (Carney, 2016; Rickford, 2016). The prominence
of the #BlackLivesMatter movement was evidenced in 2016 by #Ferguson and
#BlackLivesMatter being the first and third most used social issue hashtags,
respectively, in Twitter’s 10-year history (Anderson & Hitlin, 2016). Since
the creation of the Twitter hashtag #BlackLivesMatter, the movement has
grown to include an offline, member-led global network of greater than 40
chapters through a Black Lives Matter organization. Straddling virtual and
physical space, and with the broad goals of making visible police violence while
advocating for changing the devaluation of Black lives, the #BlackLivesMatter
hashtag is a significant case study of digital activism.
Since Trayvon Martin’s death in 2013, cases of police brutality have
remained steady, with national attention drawn to the deaths of Eric Garner,
Michael Brown, Walter Scott, Freddie Gray, Sandra Bland, and Tamir Rice,
to name a few victims. I examine the use of the #BlackLivesMatter hashtag
through the 2016 US presidential election to see how online activism sustained
and developed its political purpose through crowdsourcing gatekeeping and
framing to the networked publics that participated on #BlackLivesMatter.
Utilizing dual networked theories of gatekeeping and framing I situate the
68 Sharon Meraz
#BlackLivesMatter hashtag within the growing threats of the 2016 US pres-

idential election as it related to the political realities of Black lives. The 2016
presidential year saw the rise of White nationalism (Cox & Koebler, 2019;
Gray, 2016) and hate speech online (Guynn, 2019; Hatzipanagos, 2018;
Rainee, Anderson, & Albright, 2017; Valdivia, 2019), not just in the US but
globally (Weaver, 2018). US 2016 political publics displayed a broad loss
of trust in institutional politics (Jones, 2016; Van der Meer, 2017), shifted
toward populism and anti-establishment candidates (Galston, 2018; Norris,
2016, Shuster, 2016), and distrusted mass media as fake, especially in con-
servative politics (Allcott & Gentzkow, 2017; Benkler, Faris, Roberts, &
Zuckerman, 2017; Narayanan et al., 2018). Scholars continued to observe
in 2016 the palpable rise of fake news disseminators, or bots (automated and
semi-automated computer code), and their algorithmic role in making false
information viral and contagious (Woolley & Howard, 2016).
On a positive note, the 2016 year saw growth in the #BlackLivesMatter
movement in the UK, France, Brazil, South Africa, and Australia. The Black
Lives Matter movement aligned with the rights of other minorities, including
brown Latino publics, LBGTQ publics, and Muslims (Chan, 2016; Fadal,
2016). The #BlackLivesMatter movement was credited with changing the dis-
course on race and redefining the role of modern political protests as it related
to social media in online and offline political engagement. In May 2018,
#BlackLivesMatter appeared nearly 30 million times, which averaged more
than 17 million daily tweets (Anderson & Hitlin, 2016; Robinson, 2018).
#BlackLivesMatter, Networked Gatekeeping, and Networked

Framing
The theories of networked gatekeeping and networked framing have not
been utilized in prior studies on #BlackLivesMatter; yet, earlier research on
the #BlackLivesMatter hashtag revealed novel changes to gatekeeping and
framing practices. Through 2014 and 2015, Freelon, Mcllwain and Clark
(2016) found that protestors and supporters were able to successfully circu-
late their stories on Twitter independent of mass media gatekeepers, resulting
in the growth of six main communities: Black lives matter activists, anony-
mous hacker communities, Black entertainers and celebrities, conservatives,
mainstream news, and young Black twitter activists. Gatekeeping in the
#BlackLivesMatter movement was found to be driven by a solid young Black
demographic (Carney, 2016; Freelon, Mcllwain, & Clark, 2016). Freelon
et al. (2016) also found that participants tweeting to #BlackLivesMatter
formed a diffused, non-dense network of few reciprocal connections.
Networked Theories of #BlackLivesMatter 69
Gatekeeping, a term coined by Lewin (1947) in his study of group dynam-

ics, is defined by Shoemaker and Vos as the “culling and crafting of count-
less bits of information into a limited number of messages that reach people
every day” (2009, pg. 1). The growth of the Web in the late 1990s resulted
in scholarly dispute over the tenability of the gate metaphor as a point of
control by elites over a limited supply of information. As scholars questioned
whether elite mass media could control information with a burgeoning web
that facilitated unlimited shelf space for consumers to turn producers, Bruns
(2005, 2008) coined the term “produsage” to capture a more participatory
public. The altered information dissemination environment, driven by the
Web and social media technologies, needed to account theoretically for the
public as a central participant in the creation and distribution of informa-
tion (Bennett, 2004). Scholars of media gatekeeping have suggested revising
gatekeeping theory for a more participatory age by breaking gatekeeping into
two central processes: primary gatekeeping for information creation by mass
media and secondary gatekeeping for recirculated information dissemination
in social channels by engaged publics (Shoemaker & Vos, 2009). However,
this revised theory of elite media gatekeeping confined engaged publics to
distributing news as opposed to creating content.
This current study uses a revised theory of networked gatekeeping that
accounts for networked, multimodal flows of information among pub-
lics through the connective metrics of Web 2.0 technologies (Monge &
Contractor, 2003). The pluralized methods of Web news gathering afforded
by the sociotechnical architectures of social media technologies have rendered
traditional power hierarchies outmoded. Now, hybrid tendencies have enabled
plurality and fluidity to information gathering (Chadwick, 2013; Meraz &
Papacharissi, 2013, 2015), enabling ordinary citizens to share and trade off
power with elites in both content creation and content dissemination.
The theory of networked gatekeeping was built on Barzilai-Nahon’s
(2008, 2011) work on the networked flows of information between the
gated and the gatekeepers. Meraz and Papacharissi have defined gatekeep-
ing as “the process through which actors are crowdsourced to prominence
through the use of conversational, social practices that symbiotically con-
nect elite and crowd in the determination of information relevancy” (2015,
pg. 99). Prior studies have found that gatekeeping influence on Twitter was
ephemeral and was dependent on the issue under consideration and the
addressivity marker (mention, retweet, or via) under investigation (Meraz
& Papacharissi, 2013). This current study sought to determine the most
prominent users across the longitudinal arc of the 2016 #BlackLivesMatter
movement by asking:
70 Sharon Meraz
RQ1: Who were the prominent networked gatekeepers on #BlackLivesMatter during

the 2016 US presidential election?
Social media environments have seen the explosion of bots, defined as

automated or semi-automated algorithms/scripts that are designed to manip-
ulate public opinion by often powerful political actors (Woolley, 2016; Woolley
& Howard, 2016). Computational propaganda is the political usage of bots
in intentional, malicious ways to manipulate public opinion with one-sided
disinformation. Howard, Woolley and Calo (2018) found that US political
bots enabled surreptitious campaign coordination, solicited contributions or
votes, and violated the rules of disclosure. This study sought to determine the
prevalence of bots and suspicious accounts during the 2016 US presidential
election on the #BlackLivesMatter hashtag by asking:
RQ2: To what extent were bots prevalent as networked gatekeepers on
#BlackLivesMatter during the 2016 US presidential election?
The theory of mass media framing was designed to capture an elite-driven

process by which influentials (mass media, focal elites, politicians, and public
relations practitioners) have shaped how issues are interpreted and under-
stood by the audience. Entman (1993) defined framing as the promotion
of a “particular problem definition, causal interpretation, moral evaluation
and/or treatment condition” (pg. 52). Reese (2001) defined media fram-
ing as organizing principles that persist over time as they are socially shared.
Gamson and Modigliani (1989) defined framing through its ability to make
sense of events and suggest what issues are important to publics. Prior stud-
ies have found that the US president has an inordinate ability to frame the
media’s agenda (McCombs & Reynolds, 2002).
Unlike mass media framing, networked framing has emphasized the
altered, hybrid media ecology, the enhanced social and technical (often
dubbed sociotechnical) features of social media technologies, and the partici-
patory logic of publics empowered to engage in personalized digital mediated
activism outside elite influence. Meraz and Papacharissi reworked Entman’s
theory of media framing for a networked, participatory age, defining net-
worked framing as “a process through which particular problem definitions,
causal interpretations, moral evaluations and/or treatment recommendations
attain prominence through crowdsourcing practices” (2015, pg. 103).
However, scholars have not yet utilized networked framing as a theo-
retical foundation for #BlackLivesMatter. Using traditional framing the-
ory, Freelon, Mcllwain and Clark (2016) found that participants tweeting
to #BlackLivesMatter utilized the hashtag to frame breaking news, pin-
point police brutality, critique mass media coverage, and express support.
Oppositional framing by dissonant publics emphasized Black lawlessness and

encouraged support for police officers. These forms of resistance framing by
detractors were also found by other studies of the movement (Carney, 2016).
Prior studies have found that hashtags on Twitter enabled the framing
of issues in a bottom up, crowd-centered manner (Bruns & Burgess, 2011;
Zangerle, Gassler, & Specht, 2011), permitting ad-hoc emergent publics
to dynamically self-organize and comment on issues in a real-time manner
(Gonzalez-Ibanez, Muresan, & Wacholder, 2011; Kouloumpis, Wilson, &
Moore, 2011; Wang, Wang, & Zhu, 2013). These hashtag studies can be
leveraged toward an understanding of how networked framing processes are
enabled via hashtags. Hashtags have provided emergent interpretations along
substantive lines (e.g., #BETawards) or along sentiment dimensions (#icant-
breathe). In keeping with this prior work on hashtags as emergent frame
markers, this study asked:
RQ3: How did hashtags enable networked framing on #BlackLivesMatter during
the 2016 US presidential election?
Sampling and Analyzing the #BlackLivesMatter

Twitter Stream
This study sampled #BlackLivesMatter tweets throughout the February 2016
to November 2016 US election year using a service TweetArchivist. Across
the aforementioned time period, approximately 3 million tweets were har-
vested through hourly pulls that yielded at maximum 1500 tweets per hour.
Tweets were uploaded to a database in CSV downloaded format, and Python
scripts were written against the database to decipher prominent users and
prominent hashtags on a monthly basis. This querying of the tweet database
for users and hashtags formed the backbone of data extraction utilized to
answer the three research questions.
To determine prominent gatekeepers (RQ1), an iterative process was
employed. Tweets were collected in batches of 50,000 at a time (the CSV
rate limit of records in excel). Prominent users were determined as the top
25 per 50,000 batch of tweets; hence, several prominent users were detected
each month. From these top 25 lists, Python scripts were written to aggregate
prominent users monthly and in aggregate across the sample time period.
To decipher whether these prominent users were bots (RQ2), each promi-
nent user was checked against the automated bot detector tool Botometer.
Botometer, developed at Indiana University in a joint collaboration between
the Network Science Institute and the Center for Complex Networks and
Systems Research, has been widely utilized as a bot detector tool in prior
72 Sharon Meraz
research (Badawy, Lerman, & Ferrara, 2019; Luceri, Deb, Badawy, & Ferrera,
2019). Scores from Botometer forecasted the likelihood of each account
being a bot based on widespread markers in content and sentiment. Further
investigations into suspicious accounts (RQ2) included determining which
accounts were suspended (a sign that the account was removed due to suspi-
cious activity), blocked temporarily (usually indicated by a message that the
user has not tweeted as yet), or flagged as “does not exist” (a sign the account
was removed by either the user or Twitter).
To see whether hashtags enabled networked framing (RQ3), hashtags on
the #BlackLivesMatter stream were extracted from tweets and analyzed. The
top 25 hashtags were retrieved per batch of 50,000 tweets, resulting in several
of these top 25 lists per month. These lists were aggregated and synthesized
via Python scripting, similar to the aforementioned process to determine net-
worked gatekeepers. Prominent hashtags were clustered by their thematic
similarity via automated cluster analysis in NodeXL and further analyzed with
secondary manual content analysis checks. Hashtags were also individually
examined on Google searches via time period to determine why they were
trending. This latter process enabled the qualitative interpretation of how
hashtags functioned as networked framing devices both individually and in
bundled packages.
Who Were the Network Gatekeepers?

RQ1 sought to determine the prominent network gatekeepers. The tweet
sample identified a total of 506 prominent users across the sample time
period. Of these, 166 (32%) posted in only one month (single time point
tweeters). Users who tweeted across more than one month were more likely
to be prominent users: of the 144 users of the #BlackLivesMatter Twitter
hashtag who posted more than 100 tweets, only 14 (10%) were single time
point tweeters. Prominent network gatekeepers became visible by repeated
participation over time.
The top 25 users (see Table 5.1) revealed stark differences in posting vol-
ume between the top half and the bottom half. The most prominent Twitter
users were well ahead of the others in terms of tweet volume, with the most
prominent user posting in excess of 5000 tweets. The majority of Twitter par-
ticipants tweeted with much lower frequency; in fact, most of our sample of
506 prominent users posted fewer than 200 tweets across the sample period.
This phenomenon where only a few users are responsible for the majority
of contributions is called a power law (Barabasi, Albert, Jeong, & Bianconi,
2000; Barabasi & Albert,1999). In #BlackLivesMatter, there was a robust
Table 5.1. Top 25 Network Gatekeepers across the #BlackLivesMatter Hashtag

during 2016
Users Tweet Volume Users Tweet Volume

mikelbtko 5196 gdiciples21520 1207
SanFranMediaPR 4092 dignitasnews 1156
courtneyjahenry 3997 ogkeyser 1065
BlackjediNow 3261 WeYoungAndBlack 971
Seed_of_Israel 3019 BlackLivesToday 945
dahmacdaddy808 2221 AndreaDonteBurn 884
inspirepodcast 2181 rmilin11 767
uniqueloves 1828 chigobiker 765
zlaggy 1785 MindyYostDid 755
HotNostrilsrFun 1532 MJKramer368 751
trueblacknews 1511 XarinaFarina 735
Ironyisfunny9 1440 adami913 713
tshirtking12 1352
power law. Segmenting the top 100 users into groups of 20 users ranked
highest to lowest and running a one-way ANOVA among the 5 groups of
users in relation to their tweet volume revealed significant differences among
the 5 groups of very prominent users (f[4, 96]=27.33, p=.000). The top 20
users were significantly more likely to tweet across multiple months (M=6.45
SD=2.6) than users in the 21–40 category (M=3.9, SD=1.7), the 41–60
category (M=3.0, SD=1.3), the 61–80 category (M=2.25, SD=1.2), or the
81–100 category (M=1.5, SD=0.7). Participants in the top 21–40 category
were also more likely to post than those in the lower end of the participation
groups (61–100). Top users were top by the sheer frequency of tweeting to
the hashtag.
RQ2 sought to determine the prevalence of bots and suspicious accounts
on the #BlackLivesMatter hashtag. Of the 506 prominent unique users, 103
(almost 20%) were suspended, an alarming finding. A further 88 accounts
(17%) revealed a “does not exist” Twitter webpage, suggesting that the par-
ticipants deactivated/removed their profiles or were suspended by Twitter.
An additional 5 accounts (1%) revealed a blocked Twitter status, suggesting
temporary suspension by Twitter. In sum, 196 accounts (38%) could not be
accessed due to potential malicious activity. This malicious activity was even
more prominent among the top 100 accounts where 42 were flagged with a
status of suspended, does not exist, or blocked.
74 Sharon Meraz
Of the remaining 310 accounts, 298 were unprotected/public allowing

them to be checked for evidence of bot-like behavior. Of these 298 accounts,
41 (14% or 8% of the total sample of 506 prominent users) were flagged by
Botometer.
Networked Framing with Hashtags on #BlackLivesMatter

The 30 most popular hashtags (Table 5.2) provide a high-level snapshot of
some of the competing concerns of #BlackLivesMatter networked publics
as they sought to frame the meaning of the movement for both support-
ing and dissenting publics. A companion hashtag to the #BlackLivesMatter
movement, #blm, was one of the most utilized hashtags, as was #blacktwitter,
another large Black networked public on Twitter. The hashtag #racism was
also present through the sample period.
Competing hashtags for Bernie Sanders (#feelthebern, #berniesanders
and #bernie2016) and Hillary Clinton (#imwithher, #hillaryclinton) revealed
progressive discord among publics. Top hashtags also promoted Donald
Trump (#trump, #maga) and conservative agendas (#tcot) as well as sought
to criticize the legitimacy and power of the #BlackLivesMatter movement
(#alllivesmatter, #bluelivesmatter).
Table 5.2. Top Hashtags for the 2016 US Presidential Election on #BlackLivesMatter
Top Hashtag Frequency Top Hashtag Frequency

#blm 54705 #blackandblue 10913
#alllivesmatter 39000 #korryngaines 10734
#feelthebern 31217 #maga 9653
#altonsterling 28,197 #hillaryclinton 9289
#blacktwitter 27875 #charleskinsley 9249
#tcot 19202 #blackhistorymonth 9090
#imwithher 17034 #veteransforkaepernick 8682
#saytheirnames 16053 #berniesanders 8540
#milwaukee 14664 #sayhername 8360
#mannequuinchallenge 13329 #batonrouge 7423
#terrencecrutcher 13042 #oscars 7250
#lemonade 12531 #racism 7134
#bluelivesmatter 12288 #13th 6367
#trump 12218 #bernie2016 6340
#philandocastille 11907 #nodapl 6026
Prominent hashtags also revealed those created to address police bru-

tality and racial profiling, seen in resistance hashtags such as #mannequin-
challenge, #blackandblue, and hashtags that called attention to Black victims
such as Terrance Crutcher (#terrencecrutcher), Alton Sterling (#altonster-
ling), Charles Kinsey (#charleskinsey), and Korryn Gaines (#korryngaines).
The overarching desire of the #BlackLivesMatter movement to speak truth
and create visibility surrounding the lives lost to racism was evidenced in the
hashtags #saytheirnames and #sayhername.
These high-level popular hashtags also revealed the desire of the
#BlackLivesMatter movement to build bridges to other minority groups and
issues, as seen by their alignment with Native Americans in the North Dakota
pipeline issue (#nodapl). The movement also appeared strongly connected to
current events affecting the Black community as evidenced by hashtags such
as #oscars, #lemonade, #veteransforkaepernick, and #blackhistorymonth.
The results of the high-level framing breakdown for the top 30 hashtags
guided further analysis of how lower level hashtags, month by month, func-
tioned alongside these high-level frames to highlight the issues of concern for
the movement. This analysis revealed several networked frames that defined
and characterized the #BlackLivesMatter movement.
Networked Framing of Police Brutality: Non-stop, Heartless,

Never-Ending
The #BlackLivesMatter movement sought to elevate to high visibility the per-
vasiveness of the murder of young Black men and women. The network of
hashtags on police brutality created urgency and vigor as hashtags framed the
eruption of geographic protests about these deaths. Hashtags also served as a
reminder of past Black lives taken, enabling networked publics to revive col-
lective memory while providing updates on trials and settlements.
Calling Attention to Black Lives Lost

The sheer prominence and relentless pace of these hashtags revealed the
pervasiveness of discrimination felt against Black lives in 2016. Prominent
hashtags calling attention to Black lives lost appear every month, with the
highest prevalence occurring from July through September. August saw
19 prominent hashtags calling attention to the death of such Black men
as Alton Sterling in Baton Rouge (#altonserling, #justiceforaltonsterling,
and #batonrouge), Kendrick Johnson in Georgia (#kendrickjohnson),
and Philando Castile (#philandocastile and #falconheightsshooting) in
Minnesota, to name a few. In August 2016, 13 prominent hashtags named
76 Sharon Meraz
victims in cities such as Baltimore (Korryn Gaines: #korryngaines and #jus-

ticeforkorryngaines), Chicago (Paul ONeal: #paulonea), and Los Angeles
(Jesse Romero: #jesseromero), to name a few. In September, 17 promi-
nent hashtags called attention to the deaths of Terrence Crutcher in Tulsa,
Oklahoma (#terencecrutcher, #terencecruthcher, #terrencecrutcher, #ter-
rancecrutcher, and #tulsa) and Keith Lamont Scott in Charlotte (#keithlam-
ontscott, #keithscott, #charlotteprotest, #charlotte, #charlotteprotests,
#charlotteriots, #cltriots, and #prayforcharlotte), among others. The
names of these many victims often co-occur each month with the hashtags
#sayhername (prominent each month except April and September), #say-
hisname (prominent in July and August), or #saytheirname (prominent in
July). Alongside these hashtags are others capturing the frustration and
helplessness that supporting networked publics felt about rampant police
brutality against Black lives. These hashtags included #nojusticenopeace
(prominent in July and August), #nomoreviolence (August), #stopkillin-
gus (September through November), and #discrimination and #itstoomuch
(September).
Networked Framing of Collective Memory: Remember and

Demand Justice for Past Victims
The relentless pace of the Black lives lost in the present was highlighted
alongside the revival of collective memory of past Black victims as efforts
to seek justice for their lives remained active. Similar to calling out deaths
as they occurred, the #BlackLivesMatter public provides a running his-
tory of past deaths monthly. The anniversary of Black deaths that cap-
tured media attention and the US nation are the most prevalent through
the sample period. These include Laquan McDonald (#laquanmcdonald),
Trayvon Martin (#trayvonmartin), Freddie Gray #freddiegray, Eric Garner
(#ericgarner and #icantbreathe), and Michael Brown (#michaelbrown and
#handsupdontshoot), all of which trended in the anniversary months of
their deaths.
In addition to reviving collective memory, publics that tweeted to
#BlackLivesMatter used the community to announce settlements with the
families of Black victims. One example is that of Tamar Rice (#tamarrice) and
the $6,000,000 settlement reached with her family in April. #BlackLivesMatter
publics also utilized Twitter to mount activism campaigns to free Black lives in
jail. One such case is that of Reverend Pinkney who was arrested on charges of
election forgery in Michigan, which led to the prominence of #freerevpinkey
from May through July 2016.
Networked Framing of Building Bridges Among Diverse Minority

Publics
There were strong efforts by #BlackLivesMatter publics to connect the
Black experience with the struggle for rights by other minority publics.
In 2016, the #brownlivesmatter hashtag trended in February and April.
Strong affinities were also drawn with Muslim lives. The hashtag#mus-
limlivesmatter trended in March, August, and November. The slaying of
three young Black East African Muslim men in Indiana resulted in #our-
threeboys and#ourthreebrothers trending in March. In July, video footage
surfaced of the torturing of Aboriginal children in Australia, leading the
hashtag #handsoffaboriginalkids to trend on #BlackLivesMatter during
July and August.
The publics on #BlackLivesMatter also connected their movement to the
North Dakota Access Pipeline with hashtag #nodapl, which trended from
August through November 2016. Several focal hashtags also bound the
#BlackLivesMatter movement to the fight for rights and visibility by LGBTQ
populations, as seen in hashtags capturing a march in NYC by LGBT activists
(#bashback) and the San Francisco Pride celebration (#sfpride).
Other publics also joined forces with the #BlackLivesMatter movement to
strengthen, unify, and amplify the movement. Some of these hashtag publics
that gained prominence on the #BlackLivesMatter feed included #whitepeo-
pleforblacklives/#whitepeople4blacklives (September and October), #jews-
4blacklives (August), and #palestinians4blacklives (September).
Networked Framing of #BlackLivesMatter as a Growing Global

Movement
In 2016, the US-based #BlackLivesMatter grew to align with several other
vibrant movements globally. In late March and again in July, hashtags #blm-
totent, #toronto, #BLMTblackout, and #blmto captured the occupation of
Toronto’s square as a peaceful response to anti-Black racism and police brutal-
ity in Canada. Also in March, #prayfornigeria captured the #blacklivesmatter
publics’ response to lack of media attention to terrorist attacks in Nigeria that
month. #BlackLivesMatter publics in Australia held rallies in Melbourne in
July, captured by hashtag #melbourne. In July, the dual hashtags #blmfrance
and #adamtraore captured the death of Adam Traore in Paris. In August,
the UK #BlackLivesMatter movement marked the fifth anniversary of the
death of Mark Duggan utilizing hashtags #heathrow, #blacklivesmatteruk,
and #blmuk.
78 Sharon Meraz
Networked Framing of Current Events Significant to Black Lives

Prominent hashtags included real-world events that enabled #BlackLivesMatter
publics the space to share their opinions on significant events to their com-
munity. One big event was the release of Beyonce’s Lemonade album and
her Formation World Tour, captured by the hashtag #lemonade from April
through August 2016. The album featured the mothers of Trayvon Martin
and Michael Brown and made political statements about the Black Lives
Matter movement and Black women. Another major event was the so-called
“take a knee”movement and Colin Kaepernick’s protests, leading to hashtags
#colinkaepernick and #kaepernick trending in August, September, and into
October. Kaepernick, an NFL quarterback for the San Francisco 49ers,
remained seated during the national anthem in August as a quiet sign of
protest against the unfair treatment of Black lives in America. His protest
stance evolved to taking a knee in September. A nexus of supportive hashtags
evolved including #isupportkaepernickbecause and #isupportcolin (August
and September) and #veteransforkaepernick (September).
In 2016, the #BlackLivesMatter public also spotlighted events such as the
Grammy awards in February (#grammys, #peopleofcolor, and #movietvtech-
geeks), a BET documentary on the Black Lives Matter movement in May
(#staywokebet), the Roots miniseries remake in June (#roots), and Lousiana
floods in August (#louisianaflood, #louisiana, #batonrougeflood and #burn-
inglootingmurder). Racist KKK graffiti on a college campus resulted in
#truemu and #teamen and #easternmichigan trending in September. At
another campus in September, a student wore a gorilla mask and dangled
bananas in front of #BlackLivesMatter supporter during a protest, captured
by the hashtag #etsu.
Networked Framing of the 2016 US Presidential Election: A

Progressive Battle
The #BlackLivesMatter feed revealed a contentious battle between warring
publics in support of the Democratic rivals for the 2016 presidential nomi-
nation, Bernie Sanders and Hillary Clinton. The #BlackLivesMatter hashtag
showed strong evidence of political surveillance of primary races, as well as con-
tentious discussion of Clinton’s suitability for the nomination. Collins (2018)
argued that bots worked on behalf of Russia to sow dissent against Clinton
in support of Sanders on Clinton social media channels. Although hashtag
analysis cannot prove whether this scenario played out on #BlackLivesMatter,
I did discover strong support for Sanders, and unduly critical scrutiny of
Clinton—and no hashtags dedicated toward criticizing Sanders.
The #BlackLivesMatter hashtag provided strong networked framing by

geographic hashtags which captured the primary progressive election battle
during 2016. In February, prominent hashtags included those for the prima-
ries in South Carolina (#sccarolina, #scprimary #southcarolina, and #sc) and
Louisiana (#laprimary), the caucuses in Nevada (#nvcaucas, #nv, #nevada,
#nvdemscaucus, and #nevadacaucus), and SuperTuesday (#supertuesday). In
March, the Ohio primary (#ohioprimary and #ohprimary) took center stage,
as did the primaries in Illinois (#illinoisforbernie) and Florida (#flprimary).
April found prominent hashtags for the primaries in New York (#nyprimary,
#nyc, and #newyork), Pennsylvania (#paprimary), and Rhode Island (rhode-
island, #riprimary).
The supporter publics for Bernie Sanders and Hillary Clinton also
appeared to engage in a level of hashtag warfare on the #BlackLivesMatter
feed. Again, hashtag analysis cannot prove whether any of the sup-
porter publics for Sanders were fueled by malicious, suspicious accounts.
However, there was elevated tweeting to Sanders hashtags compared to
Clinton hashtags. Publics elevated Sanders hashtags such as #bernie (Feb
and April), #bernie2016 (February and March), #berniesanders (February
through April), #feelthebern (February through July, and September
through October), and #notmeus (February and April). A hashtag #ber-
nieorbust was elevated in March, September, and November. Clinton’s sup-
port hashtags were less prominent; her main supporter hashtags included
#hillary (February, April, and June through November), #hillary2016
(February and March), #hillaryclinton (February through April, and
August through October), and #imwithher (the entire February through
November sample period).
It is important to note that both Democratic candidates had a controver-
sial relationship with the #BlackLivesMatter movement. Sanders was repeat-
edly accosted in 2015 by #BlackLivesMatter activists for ignoring structural
racism in his issue platform. In 2016, the hashtag #neverhillary erupted in
August and October. During the August Louisiana floods, Donald Trump
accused Clinton of sleeping on the job and capitalizing on fake narratives
about her health, enabling the hashtag #whereshillary to reach prominence.
Clinton was confronted at a 2016 fundraiser about her 1996 framing of
urban Black youth as “super predators” with “no conscience, no empathy.”
Her confronter exclaimed, “I just want to know which Hillary is running for
President, the one from ‘96, ‘08, or the new Hillary?” The hashtag #which-
hillary trended in February, March, and May 2016, alongside #whichhillary
censored in February. The September 2016 hashtag #crookedhillary captured
Donald Trump’s framing of Clinton.
80 Sharon Meraz
Networked Framing of Criticism: Delegitimizing the

#BlackLivesMatter Movement
Throughout the February to November 2016 sample period, two hashtags
worked in networked concert to delegitimize the #BlackLivesMatter move-
ment. The hashtags #bluelivesmatter and #alllivesmatter were prominent
throughout the entire sample period. The prominence of these two hashtags
over time suggested a fervent, dedicated effort by contesting publics to dele-
gitimize the #BlackLivesMatter movement. The persistent presence of these
dual contesting hashtags captured the high degree of conflict and struggle
that supporting publics faced in seeking to surmount the obstacle of racism.
The #BlackLivesMatter hashtag, designed to provide a safe space for building
the movement, did anything but that. Nevertheless, these supporting publics
persisted by the overwhelming usage of the community to document police
brutality against Black lives.
Networking Framing of the Trump Campaign as Trolling

#BlackLivesMatter Publics
Alongside the delegitimizing hashtags #alllivesmatter and #bluelivesmatter
was a plethora of Trump supporter hashtags that injected noise, polluted
progressive agendas, and destabilized the platform of the #BlackLivesMatter
movement. Given Trump’s endorsement by the KKK, and his refusal to
take a stand against White nationalism or support Black activists’ agendas, it
was clear that the #BlackLivesMatter movement was largely antithetical to a
Trump presidency.
The most prevalent of these polluting or noise injecting hashtags in 2016
were #alwaystrump and #trumprally (March), #blacksfortrump and #lati-
noswithTrump (October and November), #donaldtrump (March, September,
and November), #trump (March and May through November), #trump2016
(March and May through September), #trumppence (August, September, and
November), and #maga (May through November). The presence of these tags
provided a climate of antagonism within the #BlackLivesMatter movement.
In addition to these Trump supportive tags, the hacking of Hillary
Clinton’s campaign chairman John Podesta’s email account and the associ-
ated wikileaks email daily dumps took hold of the #BlackLivesMatter com-
munity in October and November. In October, prominent hashtags included
#podestaemails, #wikileaks, #isis, #podestaemails9, and #podestaemails13.
The movement also saw the entrance of well-documented Russian troll
accounts using prominent hashtags such as #aseatatthetable, #khirzkhan, and
#blacktolive.
The Power of Networks to Create Meaning on

#BlackLivesMatter
An analysis of the #BlackLivesMatter hashtag during the 2016 US presi-
dential election revealed a very engaged, supportive networked public eager
to change the political realities of Black lives disadvantaged by discrimina-
tion and killing. This hashtag, fed by the networked actions of engaged
publics, sought to raise visibility of Black lives hurt by violence through
crowdsourcing the gatekeeping and framing processes. Alongside ad hoc,
named hashtags capturing the lives of those lost to police brutality were
geographic hashtags highlighting the places of brutality and also sites of
protest. The #BlackLivesMatter hashtag revealed an engaged public that
struggled both virtually and face-to-face to withstand widespread discrim-
ination. The prominence of a nexus of networked gatekeepers sustained
an agenda identifying brutality, calling it out by name, fostering solutions,
expressing anxiety and frustration, and seeking the best political candidate
to help restore the rights and dignities that should be willingly afforded to
Black and minority lives.
The sociotechnical affordances wrought by this new participatory ecol-
ogy (Chadwick, 2013) make the activist engagement of networked publics
more possible than in the prior media environment solely dominated by mass
media entities. As this study revealed, networked gatekeeping permitted hun-
dreds of ordinary users the chance to contribute their voices to the strug-
gle of Black and minority lives for equality. The ability of these technologies
to harness the contributions of the mass or crowd into aggregated connec-
tive action (Bennett & Segerberg, 2013) via the glue of hashtags permitted
select hashtags to function as high-level markers for networks of associated
lower level hashtags that together created the robust issue agenda of the
#BlackLivesMatter movement. The hashtags enabled supportive publics to
build up the connective memory of past victims while highlighting new vic-
tims. These hashtags provided the ad hoc flexibility to stitch the movement
into a global narrative of shared connective struggle among Black lives as
well as other threatened minorities such as indigenous, brown, LGBTQ, and
Muslim people, among others.
The 2016 US presidential election provided a useful dynamic in which
to study the #BlackLivesMatter movement. The movement, highly political
and strongly rooted in everyday lived realities, showed dynamic competition
and struggled among hashtag networked publics seeking to demand more
from their preferred nominee. However, the openness of the platform also
enabled malicious political publics to enter and perhaps destabilize agendas.
82 Sharon Meraz
It remains unclear how many #BlackLivesMatter publics truly supported

Bernie Sanders over Hillary Clinton. Post-election data revealed that 89% of
Black voters voted for Clinton and 8% for Trump (Roper Center for Public
Opinion Research, 2016). The negative hashtags that emerged as containers
for networked framing against Clinton by networked publics could have been
bot driven: about 46% of leading user accounts in this study’s sample were
suspended, blocked, or bot-like in behavior.
The openness of Twitter also enabled networked publics to develop
destabilizing agendas against the #BlackLivesMatter movement through
promoting the criticism hashtags #alllivesmatter and #bluelivesmatter. The
presence of these hashtags revealed the difficulty that social movements
such as #BlackLivesMatter face when grounded in social media technolo-
gies. Twitter’s affordance of openness and flexibility has permitted both
supporting and dissenting publics the opportunity to engage in warfare and
counterframing online (Meraz & Papacharissi, 2013, 2015). Alongside the
two dominant counternarrative hashtags was a network of hashtags support-
ing Trump. Given that Trump’s presidential campaign gained few endorse-
ments from mass media, but was supported by the KKK, White nationalists,
and online hate groups, it was apparent that the Trump supporter hashtags
worked in concert with #alllivesmatter and #bluelivesmatter to delegitimize
the #BlackLivesMatter movement.
The prevalence of suspended accounts documented the growth of
bots and computational propaganda in this study’s political landscape.
Computational propaganda has been on the rise globally (Woolley, 2016),
and the #BlackLivesMatter movement was not immune. In fact, it can be
argued that movements seeking to dismantle racism and reorder society
through abolishing privilege and discrimination face ever greater threats by
trolls, bots, and other forms of computational propaganda. Future studies
are advised to do a deeper examination of the content that these suspicious,
deleted, blocked, and bot-like accounts inject into activist movements in
order to fully understand the difficulty that publics seeking equality face in
the online social media landscape.
References
Allcott, H., & Gentzkow, M. (2017). Social media and fake news in the 2016 election.
Journal of Economic Perspectives, 31(2), 211–236.
Anderson, M., & Hitlin, P. (2016). The hashtag #BlackLivesMatter emerges: Social activ-
ism on Twitter. Pew Research Center. Retrieved from https://www.pewinternet.
org/2016/08/15/the-hashtag-blacklivesmatter-emerges-social-activism-on-twitter/
Badawy, A., Lerman, K., & Ferrara, E. (2019). Who falls for online political manipulation?
In Companion Proceedings of the 2019 World Wide Web Conference, San Francisco, CA
(pp. 162–168).
Barabasi, A. L., & Albert, R. (1999). Emergence of scaling in random networks. Science,
286(5439), 509–512.
Barabasi, A. L., Albert, R., Jeong, H., & Bianconi, G. (2000). Power law distribution of
the World Wide Web. Science, 287(5461), 2115.
Barzilai-Nahon, K. (2008). Toward a theory of network gatekeeping: A framework for
exploring information control. Journal of the American Society for Information Science
and Technology, 59, 1493–1512.
Barzilai-Nahon, K. (2011). Fuzziness of inclusion/exclusion in networks. International
Journal of Communication, 5, 756–772.
Benkler, Y., Faris, R., Roberts, H., & Zuckerman, E. (2017). Study: Breitbart-led right
wing media ecosystem altered broader media agenda. CJR. Retrieved from https://
www.cjr.org/analysis/breitbart-media-trump-harvard-study.php
Bennett, L. (2004). Gatekeeping and press government relations: A multi-gated model
of news construction. In L. Kaid (Ed.), The handbook of political communication
(pp. 283–313.) Mahwah, NJ: Lawrence Erlbaum Associates.
Bennett, L., & Segerberg, A. (2013). The logic of connective action: Digital media and the
personalization of contentious politics. Cambridge, UK: Cambridge University Press.
Bruns, A. (2005). Gatewatching, not gatekeeping: Collaborative online news production.
Bruns, A., & Burgess, J. (2011). Twitter hashtags from ad hoc to calculated publics. In
N. Rambukkana (Ed.), Hashtag publics: The power and politics of discursive networks
(pp. 13–28). New York, NY: Peter Lang
Carney, N. (2016). All lives matter, but so does race: Black Lives Matter and the evolving
role of social media. Humanity & Society, 40(2), 180–199.
Chadwick, A. (2013). The hybrid media system: Politics and power. London, UK: Oxford
University Press.
Chan, S. (2016, August 23). Black Lives Matter activists stage protests in Britain. The
New York Times. Retrieved from https://www.nytimes.com/2016/08/06/world/
europe/black-lives-matter-demonstrations-britain.html
Collins, M. (2018, February 17). Indictment: Russians also tried to help Bernie Sanders,
Jill Stein, presidential campaigns. USA Today. Retrieved from https://www.usatoday.
com/story/news/politics/2018/02/17/indictment-russians-also-tried-help-ber-
nie-sanders-jill-stein-presidential-campaigns/348051002/
Cox, J., & Koebler, J. (2019). Facebook bans white nationalism and white sepa-
ratism. Motherboard. Retrieved from https://motherboard.vice.com/en_us/
article/nexpbx/facebook-bans-white-nationalism-and-white-separatism?utm_medi-
84 Sharon Meraz
um=vicenewsfacebook&fbclid=IwAR0CBtcqzCVm29bIJh9JkafyX_UJuylnYFFuas_
dpEIdaQ0RuUoOq6XA7wI
Entman, R. (1993). Framing: Toward clarification of a fractured paradigm. Journal of
Communication, 43(4), 51–58.
Fadal, H. (2016). #BlackLivesMatter goes global. NBC News. Retrieved from https://
www.nbcnews.com/news/nbcblk/blacklivesmatter-goes-global-n618156
Freelon, D., McIlwain, C. D., & Clark, M. D. (2016). Beyond the hashtags: #Ferguson,
#Blacklivesmatter and the online struggle for offline justice. Center for
Media & Social Impact. Retrieved from http://cmsimpact.org/resource/
beyond-hashtags-ferguson-blacklivesmatter-online-struggle-offline-justice/
Galston, W. A. (2018). The rise of European populism and the collapse of the center
left. Brookings. Retrieved from https://www.brookings.edu/blog/order-from-
chaos/2018/03/08/the-rise-of-european-populism-and-the-collapse-of-the-center-
left/
Gamson, W. A., & Modigliani, A. (1989). Media discourse and public opinion on nuclear
power: A constructionist approach. American Journal of Sociology, 95, 1–37.
González-Ibáñez, R., Muresan, S., & Wacholder, N. (2011). Identifying sarcasm in
Twitter: A closer look. In HLT ‘11 Proceedings of the 49th Annual Meeting of the
Association for Computational Linguistics: Human Language Technologies: Short
papers—Volume 2 (pp. 581–586). Stroudsburg, PA: Association for Computational
Linguistics.
Gray, R. (2016). Trump leads a global nationalist wave. Buzzfeed News. Retrieved from
https://www.buzzfeednews.com/article/rosiegray/trump-leads-a-global-national-
ist-wave#.xeGPyENXxX.
Guynn, J. (2019, February 13). If you’ve been harassed online, you are not alone. More than
half of Americans say they’ve experienced hate. USA Today. Retrieved from https://
www.usatoday.com/story/news/2019/02/13/study-most-americans-have-been-
targeted-hateful-speech-online/2846987002/
Hatzipanagos, R. (2018, November 30). How online hate speech is fueling real life vio-
lence. The Washington Post. Retrieved from https://www.washingtonpost.com/
nation/2018/11/30/how-online-hate-speech-is-fueling-real-life-violence
Howard, P. N., Woolley, S., & Calo, R. (2018). Algorithms, bots, and political com-
munication in the US 2016 election: The challenge of automated political commu-
nication for election law and administration. Journal of Information Technology &
Politics, 15(2), 1–13.
Jones, J. M. (2016). Americans’ trust in political leaders, public at new lows. Gallup.
Retrieved from https://news.gallup.com/poll/195716/americans-trust-politi-
cal-leaders-public-new-lows.aspx.
Kouloumpis, E., Wilson, T., & Moore, J. (2011). Twitter sentiment analysis: The good,
the bad, and the OMG! In Proceedings of the fifth international conference on weblogs
and social media. Menlo Park, CA: The AAAI Press.
Lewin, K. (1947). Group decision and social change. In T. M. Newcomb & E. L. Hartley,
(Eds.), Readings in social psychology (pp. 197–211). New York, NY: Holt.
Luceri, L., Deb, A., Badawy, A., & Ferrara, E. (2019). Red bots do it better: Comparative
analysis of social bot partisan behavior. In Companion Proceedings of the 2019 World
Wide Web Conference, pp. 1007–1012.
McCombs, M. E., & Reynolds, A. (2002). News influence on our pictures in the world.
In J. Bryant & D. Zillmann (Eds.), Media effects: Advances in theory and research
(pp. 1–16). Hillsdale, NJ: Lawrence Erlbaum Associates.
Meraz, S., & Papacharissi, Z. (2013). Networked gatekeeping and networked framing on
#Egypt. The International Journal of Press/Politics, 18, 138–166.
Meraz, S., & Papacharissi, Z. (2015). Networked framing and gatekeeping. In T. Witschge,
C. W. Anderson, D. Domingo, & A. Hermida (Eds.), The Sage handbook of digital
journalism (pp. 95–112). London, UK: Sage.
Monge, P. R., & Contractor, N. (2003). Theories of communication networks. New York,
NY: Oxford University Press.
Narayanan, V., Barash, V., Kelly, J., Kollanyi, B., Maria-Neudert, L. M., & Howard, P. N.
(2018). Polarization, partisanship and junk news consumption over social media in
the US. The Computational Propaganda Project. Retrieved from http://comprop.oii.
ox.ac.uk/research/polarization-partisanship-and-junk-news/
Norris, P. (2016, March 11). It’s not just Trump: Authoritarian populism is rising across
the west and here’s why. The Washington Post. Retrieved from https://www.washing-
tonpost.com/news/monkey-cage/wp/2016/03/11/its-not-just-trump-authoritar-
ian-populism-is-rising-across-the-west-heres-why/?utm_term=.95ff05c83ffd
Rainee, L., Anderson, J., & Albright, J. (2017). The future of free speech, trolls, ano-
nymity, and fake news online. Pew Research Center. Retrieved from https://
www.pewinternet.org/2017/03/29/the-future-of-free-speech-trolls-anonymity-
and-fake-news-online/
Reese, S. (2001). Framing public life: A bridging model for media research. In S. Reese,
O. H. Gandy, Jr., & A. E. Grant (Eds.), Framing public life: Perspectives on media in
our understanding of the social world (pp. 7–31.) Mahwah, NJ: Lawrence Erlbaum
Associates.
Rickford, R. (2016). Black Lives Matter: Toward a modern practice of mass struggle. New
Labor Forum, 25(1), 34–42.
Robinson, A. (2018). After 5 years, Black Lives Matter inspires new protest move-
ments. ABC News. Retrieved from https://abcnews.go.com/Politics/
years-black-lives-matter-inspires-protest-movements/story?id=56702439
Roper Center for Public Opinion Research. (2016). How groups voted in 2016. Retrieved
from https://ropercenter.cornell.edu/how-groups-voted-2016.
Shoemaker, P., & Vos, T. (2009). Gatekeeping theory. New York, NY: Routledge.
Shuster, S. (2016, December 19). The populists. Time. Retrieved from http://time.com/
time-person-of-the-year-populism/
86 Sharon Meraz
Valdivia, A. G. (2019). EU and IT companies make significant progress against ille-

gal hate speech online. Forbes. Retrieved from https://www.forbes.com/
sites/anagarciavaldivia/2019/02/05/eu-and-it-companies-make-significant-
progress-against-illegal-hate-speech-online/
Van der Meer, T. W. G. (2017). Political trust and the “crisis of democracy.” Political Behavior.
Advance online publication DOI: 10.1093/acrefore/9780190228637.013.77
Wang, C. J., Wang, P. P., & Zhu, J. J. H. (2013). Discussing occupy wall street on
Twitter: Longitudinal network analysis of equality, emotion, and stability of public
discussion. Cyberpsychology, Behavior, and Social Networking, 16, 679–685.
Weaver, M. (2018, October 16). Hate crime surged linked to Brexit and the 2017 ter-
rorist attacks. The Guardian. Retrieved from https://www.theguardian.com/
society/2018/oct/16/hate-crime-brexit-terrorist-attacks-england-wales
Woolley, S. (2016). Automating power: social bot interference in global politics. First
Monday, 21(4). Retrieved from https://firstmonday.org/article/view/6161/5300.
Woolley, S., & Howard, P. N. (2016). Political communication, computational propa-
ganda, and automated agents. International Journal of Communication, 10. Retrieved
from https://ijoc.org/index.php/ijoc/article/view/6298/1809
Zangerle, E., Gassler, W., & Specht, G. (2011). Using tag recommendations to homoge-
nize folksonomies in microblogging environments. Lecture Notes in Computer Science,
6984, 113–126.
6.
The Potential of Social Media Groups
to Afford Users a Voice
James Ngetha Gachau
One of my major mentors Paulo Freire, the Brazilian educator, always says
that it is dialogue that is the true act of love between two subjects, and points
out again and again, drawing on Che Guevara and others, that there can be
no revolution without love.
—bell hooks, 1991, p. 3–4, emphasis added
The focus of this chapter is communication, dialogue, and voice. For people
to have “a responsible share according to capacity in forming and directing
the activities of the groups to which [they] belong” (Dewey, 1954, p. 147),
they must be able “to indicate the advantageous and the harmful, and there-
fore also the right and the wrong” (Aristotle, 1932, p. 11), so as to form
“an assembly in which all may hear and speak” (Shepherd & Rothenbuhler,
2000, p. 2).
A voice with which to do all this is necessary. Using ethnographic in-depth
interviewing, participant observation, and analysis of posts, I explore how and
to what extent social media in general, and Facebook groups in particular, can
provide active members with a voice that articulates their experiences, hopes,
fears, desires, needs, accomplishments, and vision. I ask the following overar-
ching research question:
Can Facebook groups oppose the hegemonic voice of the public sphere in which they
are located, and if so, how?
What is Voice?
I use the term “voice” to mean “the ability to speak on one’s own behalf,”
as explicated by Stanley Cavell (1979, p. 27). The importance of speech in
democracy was well captured in Cavell’s observation that “to speak for oneself
politically is to speak for the others with whom you consent to association,
88 James Ngetha Gachau
and it is to consent to be spoken for by them—not as a parent speaks for you,

i.e., instead of you but as someone in mutuality speaks for you, i.e., speaks
your mind” (Cavell, 1979, p. 27, emphasis added).
David Owen further argued that speaking for others does not always have
to be in assent with them. “To dissent is still to exercise one’s political voice
and, thereby, to speak for others,” (1999, p. 587). In other words, we must
be willing to listen to those who are different from us, so that we may be
in genuine community with them. Democracy is continually negotiated and
renegotiated, and, as such, it requires that its partners in interaction question
whether the partnership remains mutually beneficial. If one party feels short-
changed, then she must speak up, she must raise her political voice, not only
for herself but for the general good of the polis.
Thus, to have a voice is to be able to speak for oneself as a mature auton-
omous adult, and to have a political voice is to be able to speak with and for
others with whom one is in community.
To operationalize this definition of voice, I employ Nick Couldry’s con-
cept of voice as “the practice of giving an account of oneself, and the imme-
diate conditions and qualities of that process” (2010, p. 3). By this, he meant
the telling of a story, a narrative, about one’s life and one’s experience of the
world. Neoliberalism, however, imposes a politics that devalues voice and
reduces everything to market principles. Couldry rejoined that neoliberal-
ism is hegemonic for this precise reason. All our stories, in other words, are
supposed to be reducible to the rationale of the free market economy, which
renders us voiceless, because we cannot tell our own unique life stories unless
they toe the line of the market.
In opposition to this totalizing effect of the market hegemony, Couldry
proposed that for voice to serve as the “skills which are necessary for anyone
to give an account of themselves and others at a certain depth and with a
certain freedom” (1996, p. 322), five general principles are required. First,
voice is socially grounded, in the sense conveyed by Alasdair MacIntyre when
he wrote that “the narrative of anyone’s life is part of an interlocking set
of narratives” (2007, p. 218). Second, voice is a form of reflexive agency, by
which Couldry meant the people telling their life stories do so in Dewey’s
(1954) conception of democracy as social cooperation; we need to recog-
nize that the narratives we make of our lives are our own (agency), and
at the same time a product of our interaction (reflexive) with others, over
time. Third, voice is an embodied process, which means that every voice is
uniquely found in the concrete material life of the person telling the story
of her life. Thus, although voice arises from social processes, it must be
embodied in a particular individual life, with all its internal and external
Subaltern Social Media Groups and Users’ Voice 89
diversity, with all the contingencies peculiar to the individual’s life history.
Fourth, voice requires a material form which may be individual, collective, or
distributed. Couldry gave W. E. B. Du Bois’s example of the double con-
sciousness of “the American Negroes” (Du Bois, 1903, p. 11), who must
see themselves through the eyes of White others, to illustrate how voice
takes a collective material form. It can also take a distributed form when, for
example, it occurs as a process generated across networks, especially such
online networks as the Facebook groups in this study. Finally, voice can be
denied by voice-denying rationalities, such as those of the market economy
or, in extreme cases, those of totalitarian regimes, brought into the worst
form by Nazism, which organized “resources on the explicit basis that some
individuals’ voice and life had no value” (Couldry, 2010, p. 10).
I use these five principles from Couldry to explore how three Facebook
groups give their members tools to tell the stories of their lives in a more
authentic fashion than would otherwise be allowed by the dominant public
sphere in which they are located.
A Brief Description of the Facebook Groups

The three Facebook groups in this study are Freethinkers Initiative Kenya
(FIKA), Pan-African Network (PAN), and Women Without Religion (WWR).
Freethinkers Initiative Kenya (FIKA)

In 2011, a group of Kenyan friends came together to challenge the claims of
a self-styled prophet who asked people to give their belongings to his church
because the world was slated to end that July. These friends were incensed by
con artists who donned the mantle of “Man of God” and then proceeded to
fleece their followers in the name of getting blessings in heaven. The friends
created a website and a Facebook page to disseminate their anti-religion
views. They called this initiative Freethinkers Initiative Kenya and adopted
the acronym FIKA.
Kenya is a highly religious society (in 2012, WIN-Gallup International
ranked Kenya the 8th most religious population in the world and 2nd most
religious in Africa, behind only Ghana). FIKA questions the religious and
other taken-for-granted claims of the dominant public. Such claims are
mainly about politics, culture, and science. In a predominantly religious envi-
ronment, atheists, agnostics, secular humanists, and non-religious people in
general tend to occupy a subaltern position that reports experiences of exclu-
sion, discrimination, and even outright persecution.
Pan-African Network (PAN)
According to its Facebook wall, PAN “is a Worldwide Platform promoting the
perpetual advancement of Global Africans (Africans/African descendant peo-
ple/African Diaspora) and the African continent. … dedicated to finding lasting
solutions to situations affecting Global Africans, and … tackle[s]‌every situation
with this intent” (PAN, 2018). The majority of the content posted by group
members highlights news and historical facts often neglected by the mainstream
media and educational and cultural institutions. PAN members perceive the
entire world where Black people are located as the mainstream dominant public;
they post content that counters the opinions prevalent in that general public.
Women Without Religion (WWR)1

On its Facebook wall, WWR describes itself as follows: “This group is for
Women Without Religion or anyone who is interested in the lives and per-
spectives of Women Without Religion. We are an atheist feminist group and
eschew all unsubstantiated supernatural claims, not just deities” (WWR,
2018). Annie Chant, the principal administrator of the group, told me she
formed the group in 2012 because all the other atheist groups she was a
member of were dominated by men who often belittled women members or
made explicitly misogynistic comments and jokes. Chant is Australian, as are
most of the group’s administrators.
WWR opposes the perceived oppression of Judaism, Christianity, and
Islam. The members interviewed for this study indicated that the group acts
as a forum where they can articulate discourses that counter the opinions held
in their home societies—opinions they deem misogynistic and a product of
the three major world religions.
The underlying premise of this chapter is that these three Facebook
groups are affective counterpublic communities that bring together individu-
als with identities opposed to the mainstream public opinion of their societies.
Using the five principles of voice enumerated by Couldry, I explore whether
and how members of these groups use the affordances of social media to cre-
ate a voice for themselves qua members of the groups.
How Does Voice Manifest in the Three Facebook Groups?

Voice as Socially Grounded
Couldry argued that voice depends “above all [on] the shared resources of
material life” (2010, p. 9). These resources include both language and the
status that allows others to recognize one as having a story that is inextricably
intertwined with those of others. In this study, I found that members of each
group had their stories grounded in the language and statuses of the other
members. The posts and interviews I analyzed revealed that the discourse
engaged in by the group members was intimately related to their sense of
community with other group members.
For example, Tracy Sherman2, a 38-year-old woman and graduate stu-
dent, was born and raised in Florida in a “fundamentalist Baptist” family. The
man was always the head of the house, and all the wife could do was pray that
he was always right, even when he was objectively wrong. Tracy felt she lacked
status as a woman in her family, so she joined the US Army. She thought she
would fit right in, because of her strict and regimented upbringing, and saw
the Army as “an opportunity to get out of a world where women were at the
very bottom of the totem pole.” She had hoped to find a voice as a soldier
in a socially shared place and resource; her life’s story, her voice, was socially
grounded in order and discipline. She had also hoped that her status would
be elevated and that she would be equal with the men alongside whom she
served. But this was not the case:
I was disappointed and left after two years. There were many incidences of rape,
sexual discrimination, and when I got pregnant, I sunk even lower on the totem
pole because the epitome of a soldier is a “man’s man.” A pregnant woman can-
not in any way be that, and I was made a complete outcast.
To Couldry, voice depends on being recognized by others as having a

status that allows one to give an account of oneself. Sherman did not gain
that recognition in her family or in the Army. In WWR, however, she did.
She found like-minded people who posted about ideas and issues that were
important to her; she even compared her participation in this atheist-identify-
ing group to that of a religious community:
… you are there with your sisters and brothers, and they share your “faith”. It is
reinforcement and validation of my beliefs and who I am as a person; it is a truly
religious experience. We feed each other intellectual things to grow on; we find
out about the world through each other; we get to learn more about the world,
seeing that the group is not US-centric, unlike many other such Facebook groups.
WWR thus allowed Sherman the voice she had been denied by her reli-
gious family and the misogynistic army.
Voice as a Form of Reflexive Agency

Couldry used John Dewey’s vision of democracy as social cooperation to
explore what forms a post-neoliberal politics may take. As far back as 1918,
Dewey had cautioned that democracy should not be identified with economic
individualism as the essence of freedom of action (Dewey, 1954). Instead,

Couldry wrote, Dewey saw freedom as grounded socially in the human expe-
rience of “communicative (not merely economic) exchange through which
individuals orient themselves to the world” (2010, p. 133). These exchanges
are composed of narratives that people construct about their lives to forge
coherent identities.
The narratives our voice take in this communicative exchange, Couldry
explained, involve our taking responsibility for them, disclosing who we are as
subjects. They do not happen to us, and they are not “random babblings that
emerge, unaccountably, from our mouths, hands and gestures” (p. 8). We
create them in response to our material and social world, to make meaning of
our lives and to orient ourselves to the world. They are thus a form of agency.
Because we recognize ourselves as the agents of our voice, as the narrators
of our stories, voice is reflexive. Through voice, we reflect upon the stories
we create and tell, examining how those stories cohere over space and time,
between ourselves as individuals, and between us and our social surroundings.
The members of FIKA, PAN, and WWR proved acutely aware of how their
discursive interactions were in material and social response to their surround-
ing worlds.
For example, on January 1, 2017, one of the most active members of
FIKA posted a meme that showed a young girl asking why God seemed
to favor certain American football teams and not others, and whether the
teams that lost football matches were “starving as if they were from Africa or
something.”
The third commenter on the post asked the original poster (OP) what
he was doing now that he knew there were starving people in Africa. This
comment generated 71 replies, the majority of which were between Daniel
Ng’ang’a, one of the most vocal apologists for theism, and Muchiri Kabugi,
an atheist member I interviewed in October 2016.
Kabugi asked, “Well isn’t that God’s work? How can a mere human make
it rain in drought areas?” Ng’ang’a told Kabugi to compare the accomplish-
ments of those who believed in God to those who did not; he said that belief
in God had advanced human rights and science, whereas atheism had not
only not done anything for the world; it had also “impeded progress in every
direction.” In turn, Kabugi asked Ng’ang’a to provide evidence for his claim,
and offered a chart that illustrated the Middle Ages in European history as a
deep trough in which the world languished in darkness and ignorance, and
presenting the ancient world of Greece and Rome, and that following the
Enlightenment, as peaks of development bathed in light and progress. Kabugi
used this chart as a rebuttal to Ng’ang’a.
Ng’ang’a responded to this by saying:

Hm … that’s a colourful chart, did you draw it yourself? Incidentally, it has woe-
ful historical inaccuracies. The “Dark Ages” were actually just the Middle Ages,
a relatively peaceful time in which universities like Cambridge and Oxford were
founded. Incidentally, universities like Harvard, Yale, Princeton were founded by
Christians. Spectacles and mechanical clocks were also invented in your “Dark
Ages”. Foundations for modern science were laid in this period by priests in
monasteries who believed they were studying the handiwork of God.
Kabugi, in response to Ng’ang’a’s observation, noted that the only

reason science came from the church was because no other institution was
involved in the generation of knowledge. Furthermore, he added, most sci-
entists of that period greatly doubted the existence of God but had to main-
tain the appearance that they were theists to ensure funding, and, even more
importantly, to avoid being burned at the stake. To this, Ng’ang’a asked
for historical evidence, calling Kabugi’s statement an assertion fallacy. He
added that history shows many scientists attributed their success to divine
inspiration.
This exchange illustrates the reflexive agency of the two FIKA mem-
bers. The atheist supported the OP that belief in God is ridiculous because it
requires suspension of effective human action, whereas the Christian main-
tained that effective human action is possible only through divine inspiration.
FIKA was therefore a fertile social ground for the creation of effective voices
for its varied members; although at first blush it seemed a group dedicated
to denigrating the claims of religion, upon further examination, it is more
complex than that. Ng’ang’a, as a Christian apologist, strengthened his theist
voice by engaging atheists on their own turf. Similarly, a 24-year-old woman
who had left the group because it insulted her faith returned after realizing
that the group had instead strengthened her faith and affirmed her voice as a
Christian who was put on this earth to “share God’s love and majesty.” Even
after the protracted effort of FIKA’s atheists to show her how daft she was for
believing in the “irrationalities” of religion, she told me that the group gave
her the opportunity to tell a story that was uniquely hers; in other words, it
gave her a voice.
Voice as an Embodied Process

Whereas voice is a product of social processes and requires a reflexive agent to
create it, Couldry (2010) argued that voice also needs to be embodied by a
particular individual. Each embodiment is unique, with a set of unrepeatable
historical circumstances. Each individual reacts to and processes the world
differently, and the account one gives of this reaction and process, which is
what Couldry refers to as the voice arising from one’s lived experiences, is
uniquely one’s own.
Couldry further explained that: “Voice as a social process involves, from
the start, both speaking and listening, that is, an act of attention that registers
the uniqueness of the other’s narrative” (Couldry, 2010, p. 9–10, emphasis
in original). Like the mother responding to the demands of an infant’s cry,
our interlocutors should respond to our unique stories for us to have a voice
in any meaningful sense.
In like manner, the members of the Facebook groups in this study engaged
in a political exchange, at the very least in affective form, when they posted on
their group walls. Their voice was embodied in these posts and the resulting
discourse. They interacted with each other in a shared “space of reciprocal
exposure” (Cavarero, 2005, p. 190).
A post in June 2017 on FIKA illustrates this embodiment of voice by the
original poster and the commenters. The OP, Salome Woods, said she had
always held a grudge against Europeans and whoever else brought her ances-
tors to America from Africa, but after a recent conversation with some friends,
she reached a different conclusion:
For anyone who is a descendant of a slave, we, well I often feel cheated of my
culture, heritage and roots. We know so little about who WE ARE. … Then
it hit me. The Europeans and Arabs are not to blame for slave trade. They’re
absolved. They were simply offered a commodity. By whom? Africans. Thank
y’all for selling us.
By expressing herself as a descendant of slaves in America, Woods embodied

the voice of the African American, whom, in popular culture, many African
immigrants to the US see as rootless and shiftless. Annette Omare, a vocal
Catholic member of FIKA, commented:
I hear this rubbish from my African American cousins …, especially high school
dropouts. I can also say that in some quarters, we of African ancestry face hos-
tility from AAs. I do not recall selling anyone, I also know my ancestors did not
sell anyone.
Emily Portman, a Kenyan woman in her mid-twenties living in Switzerland,

said, “This self-pity mentality in majority of black people really drives me
crazy.” She explained that in all of human history, the strongest societies have
always taken the weakest ones as slaves. She continued that “since we now live
in a civilized world, where people have the power to choose whatever they
want to be, black people should stop all this whining and playing the victim
card. Everyone has already moved on and no one can make you feel inferior
without your consent.”
Peter Van Volk, a White European real estate developer who emigrated
to the Kenyan South Coast, commented that because “indeed the West was
guilty of slave ownership, and in a very bad way,” he was sorry. He noted,
however, that “the story is not black and white, no pun intended.” Arabs,
he explained, were the slave catchers and suppliers, although Africans were
not without blame. However, he said that to focus on the latter would be
victim-blaming because the vast majority of the victims of the slave trade were
Africans. He went on to say:
But it is history, history is full of immoral stories, wars, corruption, exploitation,
misogyny, mass murder and so on, we all have our historical occupiers, exploiters,
abusers, but we have to live in the now, and concentrate on making the now and
the here a better place to live, we own that to our next generation, so having
this hate because of crimes that where not committed to yourself is not helping
anyone, we need to forgive and move on, we cannot blame the children for the
mistakes of the parents [no matter] how wrong and harsh those mistakes might
have been, but we should also not forget those atrocities happened, to prevent
them from ever happening again.
Van Volk embodies the voice of the European immigrant to Africa. This
quote reflects not only his perception of and apology for the wrong perpe-
trated by his ancestors, but also, more importantly, his sense of pragmatism.
He has been a long-term strong supporter of FIKA as one of the best initia-
tives Kenyans have to improve their lot in a society beleaguered by govern-
ment corruption, religious intolerance, poverty, ignorance, and disease. He is
perceived by the inner core of the group as an invaluable resource due to his
practical intelligence, although a handful of naysayers in FIKA see him as a
privileged White man who exploits the cheap labor of poor Kenyans. One of
his fiercest opponents gets into spats with him that often end with her calling
him a sexual predator who preys on beach boys3 on the Kenyan coast.
Van Volk is a thoughtful man who recognizes the role of history, without
wallowing haplessly in it. He often posts about issues that matter to him,
such as misogyny. Indeed, one of his nemeses sees him as a typical member of
the “global cabal of capitalist profiteers” who uses “fashionable fads” such as
feminism and secular humanism to detract attention from their exploitation
of underprivileged populations. I contend that Van Volk’s comment affirmed
the uniqueness of the voice he embodies. He was saying that he was not just
any privileged White man, but an individual with a unique and unrepeatable
personal history.
Voice Requires a Material Form Which May Be Individual,

Collective, or Distributed
Couldry’s fourth characteristic of voice is that it requires a material form,
which may be individual, collective, or distributed.
Individual Voice
In the last section, I illustrated how FIKA members embody voice in the
unique identity of the individual. The European member of the group, Peter
Van Volk, offers a germane example of this. As already mentioned, he is loved
by the majority of the members who see freedom in terms of individual auton-
omy, where the individual’s voice is uniquely one’s own, rather than mired in
the “superstitions” of the general sociocultural collective.
For instance, after the annulled August 8, 2017 Kenyan presidential elec-
tion, he wrote a lengthy post in which he castigated the supporters of Kenya’s
chief oppositional politician, Raila Odinga, who stood as the principal oppo-
nent to the incumbent President Uhuru Kenyatta.
The first commenter said that the post was great and asked Van Volk to
make it shareable. The second commenter mocked Van Volk, calling him the
“young wise mzungu” (Mzungu is Kiswahili for White person. Calling Van
Volk “young” was ironic, because he was one of the older members; in 2014,
he told me he was 50 years old.)
Another commenter presented a nuanced critique of Van Volk’s argument:
… I don’t think that those at the reigns are interested in elections that are imma-
nipulatable (I know the word doesn’t exist, but anyway) and it makes sense for
one in their position, but the rest of us are not in their position. It would be in
our own best interest to push for institutions that we can stand by, whether they
are for our individual interests at one time or the other or not. Participating in
an election when the referee has made it clear that ‘he’ can’t guarantee a cred-
ible outcome is fooling ourselves to put it lightly. The football team you allude
to didn’t just choose to forfeit. They tried to push for additional time to allow
the ref get ‘his’ house in order but this was not in the best interest of the ruling
power. So they failed to do so, needless to say. So you trying to talk all ‘Mutahi
Ngunyi’ doesn’t work. Your argument is lacking.
Another commenter, a woman who unabashedly opposed Uhuru

Kenyatta’s government and FIKA’s tendency toward the marginalization of
women, said, “Go comment on politics in your country. You’re so ignorant
of the dynamics here.”
Van Volk’s position arose from his commitment to free market liberal-
ism, which holds the autonomous individual as the supreme bearer of moral
agency. His post leaned heavily on this Enlightenment-era ideal of Immanuel
Kant’s bootstrap “Sapere aude!” self-liberation. Although it clearly gave

voice an individual form, it failed to recognize the collective world of Kenyan
politics.
Collective Voice
To explicate how institutionalized norms can stultify the voice of an entire
population, Couldry cited Du Bois’ concept of double consciousness.
“ ‘Race’ is a fundamental dimension of how the material conditions of voice
are shaped,” he argued (2010, p. 122). In the same vein, Dorothy Smith
(1974) introduced the idea of the bifurcated consciousness of women, which
closely resembled Du Bois’ double consciousness and Franz Fanon’s masks
of the colonized (Fanon, 1970). Women’s consciousness is bifurcated, Smith
explained, because they are acutely aware of the two worlds in which modern
industrial society is organized.
The world of our everyday lives is governed by abstract concepts
made possible by escaping the concrete world of the body. A man fits
into the world of government because society is split into a “transcen-
dental realm,” where rules, concepts, and observations are made, and an
actual realm, where “[t]‌he irrelevant birds fly away in front of the win-
dow” (Smith, 1974, p. 9). In the latter realm, “a woman … keeps house
for him, bears and cares for his children, washes his clothes, looks after
him when he is sick and generally provides for the logistics of his bodily
existence” (p. 10). Men who succeed in this world become alienated from
and lose consciousness of the concrete world, while the women who take
care of that concrete world remain acutely aware of how they must live
and work in this world for the sake of that other one, hence their bifur-
cated consciousness.
Although the world Smith described is anachronistic in the 21st cen-
tury, the content posted and shared on WWR showed that the subordi-
nation of women is still alive and well. Annie Chant, the group’s chief
admin, for instance, posted an article on June 20, 2017 from The Sydney
Morning Herald about the reactions to another article in which “[m]‌other
after mother shared their stories, of thwarted expectation, of muscle-sap-
ping fatigue and a life lost they could never recover” (Gray, 2017). The
article explained that pregnancy and motherhood often lead to overworked
and underpaid women—overworked because they must take care of their
children and their homes, underpaid because employers expect them to be
less productive in the workplace. “Anyone shocked by women who regret
motherhood,” the article concluded, “isn’t listening to women” (emphasis
added).
Chant’s post was one of many that spoke about women in general and
irreligious women in particular. Of the groups analyzed for this study, WWR
most represented a collective voice.
Distributed Voice
Finally, Couldry (2010) argued that voice can take a distributed form, espe-
cially in “a large ‘community’ that involves many groups, organizations and
institutions, and provides multiple roles for individuals” (p. 100). The prob-
lem with distributed voice, argued Couldry, is that it is difficult to assess its
effect from the speaker to the hearer:
For me to feel that a group of which I am a member speaks for me, I must be able
to recognize my inputs in what that group says and does: if I do not, I must have
satisfactory opportunities to correct that mismatch. (2010, p. 101)
Distributed voice, as in social networks, holds the promise of horizontal

as opposed to hierarchical organization and cooperation. The problem with
this potential is that it often leads to effects that are of little consequence, as
evinced by the Occupy movements, or that foster a stronger more stringent
form of centralized power and hierarchy, as seen in Egypt after the 2011
uprising. Couldry’s argument was that because it is difficult to match the
voice of a participant in a social network to a particular effect in the material
world, distributed voice presents a problem peculiar to self-organizing social
networks, despite their potential to circumvent the hierarchical organization
of traditional party politics.
Atheists in Kenya (AIK) is a group that grew out of FIKA. Its president,
Henry Okullu, said during a Skype interview in September 2016 that he, and
a handful of other FIKA members, created AIK because they felt dissatisfied
with the impact FIKA was having on “the national front: Nil!” He agreed
that FIKA gave him a sense of community, identity, and voice, saying that the
group affirmed his ideas and showed him that he was not alone in being an
atheist humanist. He said the group’s encouragement made his views more
concrete. However, he also felt that FIKA was nothing more than a chat-
group. This is the kind of distributed voice Couldry was leery of.
Additionally, Okullu was suspended from FIKA in September 2017
because he violated the group’s rules by openly campaigning and advertis-
ing for atheism and for Raila Odinga during the 2017 presidential election.
A number of FIKA members saw this ban as hypocritical, observing that
Okullu had always been a campaigner for atheism on FIKA, and had once
been a darling of the same admins who now removed him from the group.
Indeed, a simple search of the group for his name shows him as far back as
2014 courting the national Kenyan media to shine a light on atheism, which
he maintained was held in irrational fear and superstition by the majority of

the Kenyan populace. An analysis of the posts that feature him reveals that the
controversy surrounding his name is based on his attention-seeking; many of
his opponents on FIKA called him a “publicity whore.”
To be sure, even Solomon Marshall, the chief admin and chair of the
group, told me in 2014 in Nairobi that Okullu and AIK were “all about
15 minutes of fame,” whereas Marshall hoped for a more lasting change
in Kenyan society. As the post made on December 25, 2015 below shows,
Okullu indeed loved controversy and publicity, and made no bones about it:
Whenever a story is covered by the media about Okullu, be it about my opinion
about Christmas or religious studies, I see a few folks saying that [I]‌am seeking
cheap publicity. Some atheists are of this opinion as well! They rage, rant and
throw all sorts of tuntrums (in public). I actually love publicity just to set the
record straight. I love being in the news. I love being in the media, press. I there-
fore look for publicity. And I get it. As for how anyone classifies my publicity is
really not my cup of coffee! If you don’t like cheap publicity, don’t be public.
Stay closed up! Stay cocooned! And if you can do a better job at publicity, be
public! Write articles, request to be on TV, go on Radio, brand yourself! Nobody
stops you! Am a publicity kind of guy! Really no apologies!
#publicity
I interpret this phenomenon as a manifestation of distributed voice. On

the one hand, many FIKA members, including such admins as Marshall,
expressed the fear that the group, despite its efforts to have offline effects on
Kenyan society, seemed to be no more than a chat-group where atheists came
to find respite from the overwhelming religiosity of the mainstream. This
resonates with Couldry’s caution that distributed voice is difficult to measure
and trace back to concrete persons. On the other hand, Okullu represents
the centralization of the self-same horizontal power of distributed networks.
Because no one seemed willing to embody the ideals that drew people to
FIKA in the first place, Okullu’s love for publicity made him dominate the
group’s voice. This became both his and the group’s boon and bane. When
his publicity stunts bode well for atheism and humanism, he was embraced
by the rest of the group as its spokesperson. When they did not, the group
shunned him.
Voice Can Be Denied by Voice-Denying Rationalities

Following Judith Butler (1993), Couldry observed that for anyone to be rec-
ognized as a subject, she or he must satisfy certain “norms of intelligibility.”
These norms are prescribed by the rationale of a gendered and sexualized
structure of interpersonal relations in heterosexual societies. Abject beings
are those who do not fit into these roles of feminine women and masculine
men, who are also supposed to follow sexual orientations and inclinations
that are mapped onto their bodies. Because they “do not correspond to this
preconditioned grid of gender and sexuality” (Couldry, 2010, p. 121), they
simply cannot be recognized as subjects with a voice. They cannot speak or
tell their stories, because, as Butler argued, they have no concepts or norms
on which to peg their narratives and therefore cannot be listened to by the
heteronormative culture. In fact, their abjectness lies in the fact that they are
the bodies, hence the voices, that those who matter must define themselves
as not, and must silence.
Although the media content and posts that populated all three groups’
Facebook walls spoke in opposition to voice-denying rationalities structured
against free thought and atheism (FIKA), African and Black lives and experi-
ences (PAN), and women’s rights and aspirations (WWR), PAN most vocally
represented this last of Couldry’s principles of voice.
Leopold Yizhak, one of the chief admins of PAN, said during a 2016
Skype interview that the group helped him give other members a more holis-
tic vision of what Pan-Africanism stood for. He quoted Kwame Nkrumah,
Ghana’s first president and prominent African statesman, who once said that
a Pan-Africanist is “He in whom Africa is born, not one who was born in
Africa.”
Pan-Africanism, Yizhak explained, was not synonymous with denigrating
Whites, or replacing White supremacy with Black supremacy. As an exam-
ple, he discussed how the Black Lives Matter movement had been co-opted
by #AllLivesMatter as “a diversionary tactic. #BlackLivesMatter means that
#BlackLivesAlsoMatter, not that #OnlyBlackLivesMatter.” This may be a jar-
ring thing to say, because, obviously, Africa is not limited to Blacks, and not
all Africans are Black, but PAN as a group, from its membership to its posted
content, seemed to equate being African with being Black. Riza Talbot, a
60-year-old Black American member of PAN who made Black dolls in
Harlem; Rok Roko, a Rwandan musician living in Paris; and Gaga Tracey, a
Ghanaian-born American who moved back to Ghana in her middle-age years,
all expressed their attachment to the group’s quest of advancing the Black
race in opposition to the legacy of exploitation by Europeans.
These PAN members expressed a voice in opposition to the dominant
worldview found across the globe that denies the lived experiences of Africans
in general and Black people in particular. The rationale behind this suppression
of Black voices is the same one that has for centuries justified the occupation
of indigenous lands by conquistadors at the expense of native populations,
the well-nigh complete annihilation of first-nations peoples in the Americas
and Oceania, and the neo-colonialism that plagues the African continent in
recent times. It can also be found in the practices by Black races across the
globe that illustrate a mentality which denigrates the self.
A PAN member, for example, posted a meme that highlighted this Black-
voice-denying rationality: gospel musician Kirk Franklin’s diminutive stature
is exacerbated by his bowing to kiss the late Reverend Billy Graham’s hand.
The description accompanying the meme scoffed at how Blacks almost auto-
matically kowtow to Whites; in fact, the picture has Graham’s face super-
ciliously cocked in a sneering smile, which the meme described as telling
Franklin, “There’s a good little N****r. No matter how big you people get,
you’ll always be inferior to us.”
Summing Up
This chapter asked whether Facebook groups can oppose the hegemonic
voice of the public sphere in which they are located, and if so, how? To what
extent and how do the three groups manifest the five principles identified by
Nick Couldry as “necessary for anyone to give an account of themselves and
others at a certain depth and with a certain freedom” (1996, p. 322)? The
findings suggest that a well-organized oppositional Facebook group is able
to afford its members the means by which to articulate a coherent narrative
of what their lives as members of a subaltern class mean. Further, the voice
afforded by the Facebook groups is grounded in the sociocultural milieus of
the group members qua members of the group in question.
The women in WWR, for example, share a history of being oppressed by
patriarchal religion, especially that of the Judeo-Christian tradition. Having
been brought up in households where the father was the supreme authority
and where this authority was reinforced by church doctrines and practices
that subordinated women to men, they find “religious and spiritual nourish-
ment” by “communing” with each other in a Facebook group that seeks to
challenge, or even abolish, these roles. By sharing their experiences of leaving
a religion that had oppressed them, they find a way to give an account of
themselves that is deeper and more authentic than the traditional one of their
dominant societies.
Similarly, PAN members post news stories that highlight the discrimi-
natory practices institutionalized against Blacks and other people of color,
predominantly in Western societies but also around the world. PAN therefore
serves as a newsroom of sorts, with the group members acting as reporters,
editors, presenters, and recipients of information that tell their story in ways
the mainstream media outlets of their societies do not.
And FIKA affords a means to create stories reflecting members’ own

authentic experiences, rather than the taken-for-granted assumptions of the
dominant public sphere. The meme that caricatured the Christian God as
more concerned with the outcome of a football match than with starving
children in Africa served similar functions as Michael Warner’s “monsters of
impudence” depicted in the counterpublics of drag queens and transvestites
(2002, p. 13).
This finding echoes James Carey’s (1989) model of communication as
symbolic ritual, where reading a newspaper, for example, unites the reader
with the characters in the stories she reads about. Communication in this view
is therefore cultural, rather than instrumental. Its goal is not so much to give
its recipients information necessary for running their lives as it is to maintain
their society as a cultural unit in time. The finding, therefore, places the pres-
ent work in the cultural studies field, traceable back to John Dewey and the
Chicago School. As Robert Park astutely observed:
The newspaper must continue to be the printed diary of the home community.
Marriages and divorce, crime and politics, must continue to make up the main
body of our news. Local news is the very stuff that democracy is made of. (Park,
1923, p. 278)
The posts and media content shared by the members of these groups
serve similar functions as local newspapers, aiming to create the Chicago
School’s vision of the Great Community as the bedrock of democracy. Park
noted that the newspaper evolved as an extension of the gossip found in
small villages, and the community found in the village could be found in the
cities only if newspapers “continue to tell us about ourselves” (p. 278). The
members of these Facebook groups can thus be said to be seeking to establish
a form of democracy, where even though they do not have the wherewithal
to influence actual public policy in the real world, they are at least aware of
how it affects them, and, more importantly, how it ought to serve them. By
telling themselves stories about themselves, they side-step the corporatized
mainstream media and puncture the canopy of hegemony those media try to
impose upon them.
Notes
1. In December, 2018, after I had completed data collection, the group changed its
name to Feminists Without Religion.
2. All names are pseudonyms. All quotes from the Facebook group pages are verbatim,
typos and all.
3. Kenyan “beach boys” are young men who proffer romantic liaisons with European
tourists along the sandy beaches of the coast in return for pecuniary gain.
References
Aristotle, & Rackham, H. (1932). The politics. Cambridge, MA: Harvard University Press.
Butler, J. (1993). Bodies that matter: On the discursive limits of “sex”. New York,
NY: Routledge.
Carey, J. W. (1989). Communication as culture: Essays on media and society. New York,
NY: Routledge.
Cavarero, A. (2005). For more than one voice: Toward a philosophy of vocal expression (P.
Kottman, Trans.). Stanford, CA: Stanford University Press.
Cavell, S. (1979). The claim of reason: Wittgenstein, skepticism, morality, and tragedy.
Oxford, UK: Clarendon Press.
Couldry, N. (1996). Speaking about others and speaking personally: Reflections after
Elspeth Probyn’s Sexing the Self. Cultural Studies, 10(2), 315–333.
Couldry, N. (2010). Why voice matters: Culture and politics after neoliberalism. Los
Angeles, CA: Sage.
Dewey, J. (1954). The public and its problems. Denver, CO: Swallow.
Du Bois, W. E. B. (1903). The souls of black folk: Essays and sketches. Chicago, IL: A.
G. McClurg.
Fanon, F., & Markmann, C. L. (1970). Black skin white masks. London, UK: Paladin.
Freethinkers Inititiative Kenya (FIKA): About. (2018). In Facebook. [Group]. https://
www.facebook.com/groups/freethinkerske/about/
Gray, A. (2017, June 20). Anyone shocked by women who regret motherhood isn’t
listening. The Sydney Morning Herald. https://www.smh.com.au/lifestyle/
life-and-relationships/anyone-shocked-by-women-who-regret-motherhood-isnt-lis-
tening-20170620-gwusj8.html.
hooks, b., & West, C. (1991). Breaking bread: Insurgent black intellectual life. Boston,
MA: South End Press.
MacIntyre, A. C. (2007). After virtue: A study in moral theory. Notre Dame, IN: University
of Notre Dame Press.
Owen, D. (1999). Cultural diversity and the conversation of justice Reading Cavell on
political voice and the expression of consent. Political Theory, 27(5), 579–596.
Pan-African Network (PAN): About. (2018). In Facebook. [Group]. https://www.face-
book.com/groups/587362398051938/about/
Park, R. (1923). The natural history of the newspaper. American Journal of Sociology,
29(3), 273–289.
Shepherd, G. J., & Rothenbuhler, E. W. (2000). Communication and community.
Mahwah, NJ: Routledge.
Smith, D. (1974). Women’s perspective as a radical critique of sociology. Sociological
Inquiry, 44(1), 7–13.
Warner, M. (2002). Publics and counterpublics. New York, NY: Zone Books.

WIN-Gallup International. (2012). Global index of religiosity and atheism—2012.http://
www.wingia.com/web/files/news/14/file/14.pdf.
Women Without Religion (WWR): About. (2018). In Facebook. [Group]. https://www.
facebook.com/groups/WomenWithoutReligion/about/
7.
“Glamorous factories of unpredictable
freedom”: Care, Coalition, and
Hacking Hacking
Christina Dunbar-Hester
Let’s start to work on this and see what would happen if we change the
somehow boring hackerspaces of the present into some glamorous factories
of an unpredictable freedom for all of us even those who do not fit in the
classical nerd scheme. Change the nerds. … For you and for me and the entire
human race.
—Grenzfurthner & Schneider (2009)
Hacking is an ideologically charged mode of technical and cultural produc-

tion. In the first decades of the 21st century, it casts a long shadow. Many
undertakings might now be considered variants of hacking, from modding
cars to crafting to DIY. An earlier and more bounded meaning centers around
computer programming, of course. But hackers were not always coders; their
lineage can be traced through phone phreaking and other sorts of tinker-y
attempts at technical mastery, including lock-picking, in the latter half of the
20th century.
Rather than focusing on material output—lines of computer code, picked
locks, or other artifacts—we should consider hacking as a worldview. Its
adherents assert that cultural and technical artifacts should be left open to
allow endless modification, reinterpretation, and re-fashioning toward pur-
poses beyond those for which they were originally created (Coleman, 2012;
Jordan, 2017; Kelty, 2008). This fervent commitment to openness has led to
some cultural blind spots, including a central contradiction in hacking and
open source software communities. On the one hand, these communities’
norms dictate that these communities be open to whomever wants to be
there, which has tended to perpetuate the notion that if some people are not
there, it is because they do not wish to be, or because they are unworthy;
community members historically have claimed that they are satisfied having
106 Christina Dunbar-Hester
their ranks constituted by the best and brightest who both self-select and
prove their merits (Nafus, 2012; Reagle, 2013). On the other hand, because
hackers are evangelists for their mode of technical engagement, they believe
that more people hacking will lead to the creation of more, new, and better
technologies, so they wish to increase their ranks and thereby the revolution-
ary potential of technological production in general and networked comput-
ing in particular. Growth potential in hacker communities is thus held to be
nearly unlimited: in the breathless words of the Free Software Foundation,
“If we want to make proprietary software extinct, we need everyone on the
planet to engage with free software. To get there, we need people of all gen-
ders, races, sexual orientations, and abilities leading the way. That gives the
free software movement a mandate to identify under-represented groups and
remove their barriers to access” (2012). These issues matter even outside of
hacking because peer production and self-organizing modes of production,
key features of digital cultures, are commonly assumed to have implications
for how industrial patterns may be organized in the future (Turner, 2009).
Especially when notions of peer production become hitched to matters of
social organization, there are good reasons to consider carefully the implica-
tions for social equality and egalitarian values (Kreiss, Finn, & Turner, 2011).
This chapter is not about hacking in hardware or software, but rather
about hacking hacking itself. In the early years of the 21st century, as free
software communities matured, they began to recognize that their contribu-
tor bases were overwhelmingly composed of men. A 2006 European Union
policy report revealed that fewer than 2% of free software practitioners were
women, which catalyzed attention to these matters (Nafus, Leach, & Krieger,
2006). Many hackers decided that what Grenzfurthner & Schneider (2009)
called the “classical nerd scheme”, which has tended to favor men and elites,
was insufficient to realize their goals (Dunbar-Hester, 2016; Eglash, 2002).
With increasing urgency, groups formed to support individuals defined as
“others” in open source and hacking. Significantly, the rough consensus and
running code ethos that supported practitioners’ self-organization around
technical production was reoriented to hack their communities. These volun-
taristic efforts to reconstitute open technology communities are the subject of
this chapter. In order to assess them, I draw on discussions in feminist Science
& Technology Studies (STS). I argue that focusing on the missing or under-
represented people in hacking is an extension of I-methodology, and that
people who wish to hack hacking on more generative grounds should lessen
their focus on technological production and instead emphasize care in social
relations, coupled with critical analysis.
“Unpredictable freedom”: Care, Coalition, Hacking 107
Sites and Methods

The sites of engagement with technology around which we can witness hack-
ing being hacked can be grouped together under the umbrella of open tech-
nology, especially but not limited to free and open source software.1 Free/
Libre and open source software (FLOSS) is a set of practices for the distrib-
uted collaborative creation of code that is made openly available through a
reinterpretation of copyright law; it is also an ideologically inflected mode
of production and authorship that seeks to reorient power in light of partic-
ipants’ understandings of the moral and technical possibilities presented by
the Internet (Kelty, 2008). Hackerspaces are a cognate offline phenomenon,
community workspaces where people with interest in computers, craft, and
other types of fabrication come together to socialize and collaborate. These
sites are far from monolithic, but they are more alike—bound together by a
shared (if not singular) political and technical imaginary—than they are dif-
ferent. Hacking here is about an expression of agency and not necessarily a
desire to trespass or “pwn” (defeat) (although some hacking subcultures pos-
sess this feature). Open technology broadens the ethos of FLOSS to encom-
pass artifacts beyond software, including craft and hardware (Fox, Ulgado, &
Rosner, 2015; Powell, 2012).
Hacking and FLOSS participation often take on meaning as communal
and shared actions. Anthropologist Gabriella Coleman has demonstrated that
hackers deploy a range of political stances including agnosticism and denial of
formal politics (exceeding software freedom), although implications for intel-
lectual property in particular are at least implicit and often explicit in the tech-
nical and social practices of hacking (Coleman, 2012; Kelty, 2008). Scholars
have noted that the denial of formal politics makes FLOSS an unlikely site
for gender and diversity activism, at least historically (Nafus, 2012; Reagle,
2013). But FLOSS and hacking projects are not monolithic and have matured
over time; arguably, the diversity advocacy that is the subject of this chapter
represents a turning point within the collectivities whose focus is on hacking.
The shared enthusiasm for hacking and crafting code that unites FLOSS com-
munities has collided with a realization that to consider these communities
open in an uncomplicated way is naïve. FLOSS projects and hackerspaces are
also in dialogue with the wider culture, which is awash in “women in tech”
discourses (including the high profile of Facebook COO Sheryl Sandberg’s
2013 book Lean In). The raft of open technology initiatives geared toward
changing their communities’ constitutions must be placed within this con-
text, while keeping in mind that geek politics exist along a continuum.
In this chapter, I examine diversity work in hacking. Elsewhere I have

asked, “if ‘diversity’ is the answer, what is the question?” (Dunbar-Hester,
2019). In asking this, I mean to zero in on the complex and sometimes con-
tradictory ways in which advocates articulate and operationalize their efforts.
Unlike actors, I am not seeking answers for how to get more women or
members of other underrepresented groups into hacking or technological
production. This is not to denigrate their efforts, but to enunciate that my
work as analyst is distinct from theirs. Rather, I am inquiring into the political
potentials and limits of emphasizing particular aspects of these matters (and
muting others).
Diversity advocacy is multi-sited and multi-vocal. My research methods
here are informed by an ethnographic sensibility, but lack the deep hanging
out component that is a hallmark of traditional single-site ethnographic stud-
ies (Geertz, 1998). Instead, I mirror the distributed nature of this advocacy,
conducting participation observation at a number of sites (predominantly
North American and a few European hackerspaces, fablabs, software confer-
ences, unconferences for women in open technology, corporate events, and
software training events/meetups).
One thing to note is the relevance of my own subject position and social
identity to this research. As a White, middle-class, highly educated and literate
person in North America, and native English speaker, these communities and
their conversations are relatively accessible to me and hospitable to my pres-
ence; my presence usually required little justification, although I did identify
myself as a researcher. That said, my training, expertise, and commitments
are those of the academy, specifically interpretive social science, not computer
coding, geeking or hacking, navigating NGOs or startups, or feminist activ-
ism. Of special importance is my position as a person with a feminine gender
identity. Many of these sites are closed to people who do not identify as women
(although most are explicitly genderqueer and trans*-inclusive, some required
that people identify as women “in ways that are significant” to themselves).
Fieldwork and data gathering spans approximately 2009–2016, with con-
tinuous attention to listservs and online traffic, and punctuated conference
attendance and interviewing. This period is meaningful because it saw sev-
eral feminist hackerspaces appear as well as growing attention to gender in
mainstream open source; at the same time, it is a snapshot of an unfolding
story with both a prehistory and a future that are outside the scope of the
present research. It is significant that several initiatives that became research
sites were born during this period; although this indicates that I had my fin-
ger on the pulse of a meaningful social phenomenon, it also means that the
objects of study were a moving target and hard to identify before the fact,
which creates a methodological challenge. I interviewed participants in these

activities as well as founders of hackerspaces, open source software projects,
and initiatives to promote diversity in open technology (16 semi-structured
interviews and around 10 informal interviews). And I followed much online
activity, lurking on project lists and following social media, which is valuable
because many of these efforts are coordinated and distributed across space,
even if they also include local, static components in real life such as hacker—
and maker-spaces, or project—or programming-language-based meetups.
Conferences, of course, are important not only for the ritual elements that
occur when a community comes together for a short time, but also for the
information that is transmitted within them, so I made a special effort to
attend a variety of in-person workshops and conferences. In weaving together
these threads of activity, I map the meaningful (and contested) discourses that
suffuse diversity advocacy.
Names in this account have been changed. Though some informants said
they did not mind being identified by either real name or hacker handle,
many were concerned about appearing to speak for workplaces (especially) or
voluntaristic organizations they did not feel authorized to represent. Thus,
I opted to give pseudonyms to everyone, usually selected in consultation with
the people to whom they are applied. I do not name FLOSS projects or
hackerspaces in this chapter because their communities’ inner workings can
be sensitive for participants. My goal is to provide plausible deniability and
breathing room for the real people whose activities and utterances I narrate,
rather than deep-cover anonymity. I have made the calculus that any quotes
I include are of analytical benefit and unlikely to bring harm; I weighed para-
phrasing quotes, but decided that doing so would likely distort meaningful
elements of actors’ statements.
“I think gender disparity in Free Software is probably a Bad

Thing”2
As noted above, much agitation in open technology circles began with rec-
ognizing that open technology communities skewed masculine, often heavily.
As FLOSS matured, groups were founded to address questions of identity
and representation, primarily centering on gender. These include LinuxChix
(founded ca. 1998), Debian Women (ca. 2004), Ubuntu Women (2006),
the Geek Feminism project (ca. 2008), PyLadies (from the Python computer
language community; 2011), and more.
One mundane example of this turn toward gender awareness can be seen
in the following email exchange. One person (with a masculine username)
addressed the [Womeninfreesoftware] email list, hoping to recruit women to

FLOSS projects in which he was involved:
I had a look at the projects I’m directly professionally involved in—[Project A]
and [Project B]. And, well, they’re pretty much your typical FLOSS sausage fests
[men-dominated spaces], I’m afraid. We do actually have a few women involved,
but they’re all [company] employees [who are paid to be there]; on the volunteer
side, it’s all men so far. So I’m hoping to encourage people—women in particu-
lar—reading this list to come and get involved with [Project A] and [Project B].
(Email,——to [L—] to [Womeninfreesoftware], 9/28/09; emphasis added)
This quote exemplifies a project participant reaching out to women and in

so doing, demonstrating that these projects were aware of diversity issues
and making an effort to be welcoming. Another list subscriber (with a
feminine name) replied, “Thanks, [L]‌, for taking the time to make that
bid for participants in your project. It was exactly what the world actually
needs[,] much more so than almost any other single action” (Email, [K—] to
[Womeninfreesoftware], 9/28/09; emphasis added). These quotes illustrate
the typical, mundane framing of diversity as inclusion of women in free soft-
ware projects (all post-2006 initiatives should be read in part as being sparked
by the E.U. report showing a minuscule rate of participation by women).
Reaching out to women was seen as a straightforward, and appreciated, way
to foster the inclusion diversity advocates prized.
At the same time, people immediately recognized that there were poten-
tial pitfalls lurking in these efforts to promote diversity in their communities.
One major stumbling point, which seemed all but impossible to resolve, was
how to challenge normative masculinity as a cultural default without invoking
normative femininity as its opposite. This played out in a variety of ways. On
the list where the person had lamented the “sausage fests” in his current open
source projects, other list members discussed making a logo to represent the
list itself. One list subscriber proposed, “If we took the picture of a GNU
used by FSF [Free Software Foundation] … added lipstick, eye shadow, and
mascara, replace the beard by a string of pearls, and replaced the horns by
a feminine hat, with a flower sticking up from the hat, I think that would
convey the idea” ([M—] to [Womeninfreesoftware], email, 9/24/09). The
GNU she references is the logo of a Unix-like, Linux-related operating sys-
tem, a line drawing of the antelope-like gnu, replete with chin tuft and horns.
Essentially, the subscriber proposed adorning the gnu with normative mark-
ers of femininity.
Responses to this suggestion registered immediate discomfort. One per-
son commented, “I … am not a big fan of this idea. Most women in free
software do not adhere to traditionally feminine styles of dress/grooming—I
have seen very few wearing makeup let alone pearls at free software events—
and I think this sort of appearance would be alienating to many of us” ([K—]
to [Womeninfreesoftware], email, 9/24/09). The original poster agreed with
this: “You’re right … Most of us don’t dress over-the-top feminine. I certainly
don’t,” and added that her original suggestion was intended to be a humor-
ous way of depicting women in FLOSS ([M—] to [Womeninfreesoftware],
email, 9/24/09). Posters to the list struggled with how to signal the presence
of women without falling back on representations of normative femininity
that many found “alienating.” (They also touched on race; as one commenter
wrote, “I think the gnu is more appealing than the wasp-y [White Anglo-
Saxon Protestant] noses and dainty lips [in some other ideas for a logo]”
[R—] to [Womeninfreesoftware], email, 9/24/09].) In these discussions,
FLOSS diversity advocates were caught on the horns—pun intended—of how
to register feminine presence without invoking normative femininity. Many
women in general, and geek women in particular, exhibit discomfort with
so-called “pink technology,” or marketing strategies that draw on stereotypes
of women and girls to attract women to male-dominated products and activ-
ities (Kearney, 2010); diversity advocates were hard pressed to signify women
in ways that were recognizable without being cringe-worthy.
Another immediate line of critique had to do with gender itself: many in
open technology circles were uncomfortable reifying notions of binary gen-
der and gender essentialism. On another list, one person invited list members
to a corporate-sponsored women-in-tech event called “IT’s not just for the
boys,” at which she would be speaking. Another list member wrote back,
[T]‌he language of a lot of these events and postings seems very troubling to me.
It seems fitting that [the corporate sponsor] would host an event like this,
one which simultaneously reinforces gender binaries and makes them look good
for “supporting a good cause”. If I had the chance to speak at one of these,
I would definitely talk about how not all the boys are just boys and not all the
girls are just girls and IT should be one more way we can challenge binaries of
identity that force people into gender boxes. So is this event also for male-to-
female students? Or female-to-male students? Or genderqueer students? I would
love to see more queer feminist technology events …
([C—] to [Gender Tech list], email, 5/28/12)
This list poster is a trans woman, whose first reaction to the topic was to insist
on a trans-inclusive and gender-binary-questioning approach to women in
tech discourse. Notably, neither her identity nor her comments were anom-
alous in these voluntaristic tech spaces. At events, some of which were wom-
en-only (but explicitly allowing people to identify as women, however they
wished), people commonly expressed the utility of advocating for women
even while simultaneously recognizing that woman was an essentialized cat-

egory, which they reserved the right to question. At an unconference for
women in open technology, which I attended in Washington, DC, in 2012,
one person who identified as “genderfuck” or gender-fluid also said that they
were aware that, “I have more street cred if I say I am a woman fighting
the gender gap [in tech] than if I say I’m a gender-fluid person fighting the
gender gap [so as a matter of strategy I will say I’m a woman]” (Fieldnotes,
7/11/12). Another person at the event agreed, adding: “Being in tech has
changed how I identify. I didn’t really care [about gender before] but now
I’m like, ‘I’m a woman! Women matter! We need more women!’ But if I need
to throw in the word ‘trans’ because I think the room needs to hear that, I’ll
throw in the word ‘trans’ too because I can do that.” Even community mem-
bers who were cis-identified acknowledged the complexity of the category of
feminine gender, while still often using “women” as a shorthand.
Doing a debrief after a 2018 workshop, one person emailed the Gender
Tech list to say that, “[O]‌ne of the common feedbacks from this year’s [event]
is that people did not understand the [distinction for portions of the event
that were mixed-/all-gender, versus ones that were more closed, i.e. reserved
for women, queer/non-binary, and trans* people], and [this] needs to be
more clearly communicated [in the future]. … We need to be clear that [our
event] is special and that it has its roots in a women-only project” ([Charlotte]
to [Gender Tech] list, 10/23/18, email). Charlotte’s statements reflected a
growing consensus that presenting an event as “women only” was insufficient
in an era when gender essentialism was under question, even though the event
had been a women-only space since the 1990s. A 2018 compromise allowed
cis-men to attend some of the event sessions, but limited other sessions and
all leadership and presentation roles to “women, trans, genderqueer, gen-
der-fluid, and intersex people.” However, this understandably confused a few
attendees and even some organizers. Charlotte concluded her email by sug-
gesting, “As long as the patriarchal gender imbalance persists, there will be a
need for those who are ‘other’ than cis-men to carve out a space where our
voices and needs are heard and prioritized.” This reflected her group’s up-to-
the-minute struggle with accommodating a changing notion of gender (fluid,
not fixed) while staying true to the political and practical reasons the group
had asserted the need for a women-only space in the first place. Charlotte
was also careful to include a thoughtful explication of the values of hacking,
saying that any iteration of the event should be “self-organized and autono-
mous … & adaptable into the future.” This represents, as argued above, the
hacker belief that their activities are recursively world-making, and that the
greatest good is to leave artifacts and code open for future modification. In
this case, group members agreed that, as one poster said, “the actual goal
is freedom from the undesirable attitudes which in patriarchy are hitched to
gender,” thus leaving open the possibility of further modification down the
line ([J] to [Gender Tech] list, 10/23/18, email). In other words, they are
hacking hacking, and in true hacker style, assuming it will need to be hacked
again, by themselves or by others. As Chris Kelty wrote, “[T]echniques and
design principles that are used to create software or to implement networking
protocols cannot be distinguished from ideas or principles of social and moral
order” (2005, p. 186).
I-methodology and the Challenges of/to Hacking Communities

The above efforts of diversity advocates are thoughtful and sincere. Nonetheless,
there is an irony to much of their advocacy. Some of the most nuanced ways
of addressing social difference are at odds with realistic goals for voluntaris-
tic groups brought together by the shared social imaginary of open technol-
ogy. For example, Glen, a North American White man in his thirties who
had founded a Python language–based FLOSS project renowned by diversity
advocates for being hospitable and welcoming, said in an interview, “African
Americans are underrepresented [in open technology], but more than half
of Americans are women. If we do [outreach] efforts focused on women, a
lot of the women we’ve been attracting aren’t White, [this is] increasing the
spectrum already. [Racial diversity] won’t completely take care of itself but
[outreach to women] will address most levels” (Interview, 7/3/12). Glen
is correct that African Americans are underrepresented in open source. (In
fact, Callahan, Hathaway, and Krishnamoorthy [2016] broke representation
down into FLOSS project member versus contributor and found that just
looking at members overcounted the African Americans and Hispanics3 in the
FLOSS project they studied; when looking at the ethnicities of contributors to
the project, the presence of African Americans and Hispanics was even more
limited.) And he is also obviously correct that not all women are White, and
that outreach to women may signal to members of various groups that there
is attention being paid to inclusion. At the same time, as Glen acknowledged,
to lump racial inclusion in with gender inclusion is to paper over some of the
unique features of cultural and personal histories that members of different
racial and ethnic groups have vis-à-vis technological cultures and online cul-
tures (Daniels, 2016; Nakamura, 2013; Noble & Tynes, 2016).
In particular, the experience of vulnerability is one that stands out in the
accounts of some women of color in open technology. In describing an ugly
instance of online harassment to which she was subjected in an open source
community, Helena, a North American mixed-race woman in her thirties,

said that she felt her experience was intensified as a woman of color. She said,
“I was way more scared than women not of color [would be]”, indicating that
misogynistic attacks online can also be racialized—and often are (Interview,
8/14/16) (see Banet-Weiser & Miltner, 2016, p. 171; Nakamura, 2013). In
addition, Helena’s mundane experiences in open technology communities—
as opposed to the one where she was viciously attacked—left her unsure who
to trust. She did not always feel that White women were her allies. (Similar
dynamics have arisen with women of color in the global south vis-à-vis White
feminist FLOSS collectivities whose members are largely based in the global
north; see Dunbar-Hester, 2020.) This points to a double bind for advocates
collapsing “diversity” into “gender/women,” leaving these advocates unpre-
pared to attend to how differential social statuses and vulnerabilities play out.
Although these issues are important, so far, they are not specific to open
tech communities, and largely mirror the general social experience of racial
and ethnic minorities in cultures where Whiteness is hegemonic (Daniels,
2016; De la Peña, 2010). One issue that stands out for further consider-
ation is that open technology communities are voluntaristic. Unlike insti-
tutions of higher education or workplace relations between employers and
workers, whose terms of association are formally overseen by regulation and
the courts, these sites are predicated on elective associations between mem-
bers. As cultural historian Thomas Streeter wrote, in “creat[ing] social and
organizational structures that by their design could motivate individuals to
collaborate,” engineers whose practices laid the groundwork for open source
predicated the existence of these communities on affective commitments and
the motivation to collaborate (2010, p. 105). Expressions of good will toward
others and inclusivity are laudable as values and practices, but the terms of
association in these communities are much less enforceable than in more for-
mal institutional relationships. And as Francesca Polletta (2002) noted when
considering friendship as a basis for political organizing in the women’s liber-
ation movement, these elective associations are especially vulnerable to strain
and even dissolution when personal relationships are strained.
What is occurring here exposes a tension between the sincere efforts of
diversity advocates to be inclusive, to hack hacking, and to “change the nerds”
by issuing a challenge to open technology communities as they are currently
constituted. On the one hand, these efforts represent a sincere, iterative, and
always-incomplete effort at inclusive world-making that aims to hitch a hack-
ing ethos to diversity and inclusion work. On the other, voluntaristic efforts
within hacking communities begin, naturally, with who is already there, and
the calls for collectivity formation that began in 2006 centered mainly on
women. These calls both represented and constituted the communities who
were clamoring to change the nerds. These dynamics could rightly be called
I-methodology, in which technology designers implicitly or explicitly consider
themselves as representative of the user base (Akrich, 1995; Oudshoorn,
Rommes, & Stienstra, 2004).
Feminist scholars in particular have critiqued the I-methodology for its
implicit, covert, and often unconscious bias. As Oudshoorn et al. wrote,
“user-representation techniques more often function as tools to legitimate the
design process so that designers can claim that they have taken the needs of users
into account as tools to guide technological decisions” (2004, p. 43, emphasis
added). The danger is two-fold; both the biases I-methodology introduces
and the claims to legitimation designers make on behalf of others are prob-
lematic if the goal is to produce maximally democratic technological designs.
But in the case of voluntaristic communities, I want to suggest that it
is useful to parse these dangers carefully. In open technology communities
attuned to diversity and inclusion, there is indeed great danger in claiming
to take the needs of others into account when one is primarily speaking for
oneself. Yet insisting on meeting the needs of oneself or others with whom
one is aligned in agitational coalition is potentially the most realistic way to
change communities formed around affinity for hacking and technology. In
the above cases, where community members (including men) respectfully
supported one another in calling out for more women to volunteer, and in
redefining who women are, while leaving open the possible need to revisit
these issues and redefine membership categories in the future, a reflexive
I-methodology advanced the cause of inclusive hacking, at least somewhat. But
it is important to get the analysis right. As Glen, the Python project leader,
indicated, it would not do to conflate measures that might be specifically
welcoming to women in general, with those of African American women
in particular (for example); the mixed-race woman commenting on her fear
invoked her own differential burden in the matrix of domination (Collins,
2000). At the same time, voluntaristic technological collectivities are not, in
and of themselves, responsible for overcoming the burden of ingrained struc-
tural problems including racial and economic segregation and inequity. Their
impulse to broaden participation in their ranks in reflexive and flexible ways is
laudable, but likely to be limited in scope.
Voluntarism and Always-Incomplete, Coalitional Politics

Before concluding this chapter, I will draw out some of the issues at stake in
these contestations that are often underarticulated in open technology circles.
In particular, market logics deserve scrutiny. This chapter has elevated the
voices and social world of those agitating within open technology communi-
ties, but of course their social world shades (and fragments) into other worlds.
Industry and higher education are also pushing for diversity in science,
technology, engineering, and math (STEM), including but not limited to
women (Margolis, 2010; Poster, 2008). These sectors are primarily invested
in national competitiveness, economic growth, women’s economic empow-
erment, and producing (diverse) products for a (diverse) consuming market
(and of course, these mandates cannot be cleanly separated). As of this writ-
ing, the National Center for Women & Information Technology (NCWIT)
reported that 26% of the U.S. computing workforce is women (and less than
10% is women of color: 5% Asian, 3% African American, and 1% Hispanic).
Diversity advocates often find that their messages can proceed with great
ease when the rationale for their efforts dovetails with calls for diversity in
industry spaces. In fact, the very idea of diversity holds great institutional
appeal (Ahmed, 2012). Of course, one reason for this resonance has to do
with open technology diversity advocates’ own adjacency to interpretive
frameworks that are devoted to the profit-oriented pursuit of technological
development and growth; many advocates are employed as tech workers. Yet
voluntaristic open technologists’ motivations are not identical to those of
industry players’, and they do themselves a disservice to allow their own calls
for diversity to converge with those of industry.
When extolling the virtues of diversity in tech, Christen, a 31-year-old
German, said, “I cannot buy a bigger smartphone because it won’t fit in my
pockets [as a woman]. Apple didn’t include period tracking in their health
app, [and] face recognition software regularly fails people of color” (personal
correspondence with author, 7/2/15). This is a perfect encapsulation of a
market logic being articulated by a volunteer diversity advocate: she touches
on race, gender, and consumption, but steers clear of the controversy that
surrounds, e.g., face recognition software, such as its relationship to surveil-
lance and algorithmic incursions into citizens’ compacts with states.4 And yet,
she also disclosed a more expansive notion of what is at stake for her: “I wish,
more diversity would mean for everyone, who is not a white heterosexual
able-bodied male, to finally feel normal and not like a freak … Even if it meant
just this bit of respect and humanity it would change the world” (emphasis
added). Although she invoked respect for difference and humanity at the core
of her vision for diversity in tech, product-centered explanations of the value
of diversity are always rhetorically within reach and always an easy shorthand.
But this shorthand shortchanges diversity advocates who desire to pursue
social justice (Ahmed, 2012).
Diversity advocates who seek to advance a notion for hacking hacking that
exceeds the boundaries of industry’s concern may wish to include a critical,
reflexive notion to the social problems that technological participation can
and cannot solve. Social power and technical participation are imbricated to
such a degree that they may at first glance seem interchangeable, but increas-
ing participation in technological domains is no guarantee of movement
into a more empowered social position. In fact, science and technology have
historically been sites for cultural sorting work, separating so-called “STEM
capable” people from “STEM incapable” people, according to historian of
technology Amy Slaton (2017) and others (Fouché, 2003; Harding 1995,
2016). Simply moving people from one category to another reinforces the
use of STEM as a site for this kind of problematic cultural sorting. To place
technology at the center of a project of social empowerment is uniquely chal-
lenging for these reasons.
If diversity advocates hope to see change in how power relations in tech-
nological communities are constituted, it is imperative that they not only
attend to who is there (although this is certainly important). They must ana-
lyze the role of technology in maintaining the social order of the wider soci-
ety. Paradoxical and challenging though this may be for collectivities called
into being by a shared affinity for technology, it is essential to recognize that
neither diversity nor technology can stand in for social or economic justice,
antiracism, or feminism, materially or conceptually. Further, returning to the
use of identity and the measure of one’s own experiences to articulate what
must change in open technology communities, the above vignettes illustrate
that, by its nature, I-methodology can be reflexively brought to bear to offer
partial prescriptions for changing communities. In voluntaristic communi-
ties, which are tending their own gardens even as they hope to effect more
sweeping change, it is probably better to ambivalently embrace these par-
tial, reflexive, and self-referential interventions as though they are features,
not bugs.
It is unrealistic for DIY collectivities to do much of the heavy lifting
of structural change, and this is not their shortcoming. Diversity advocacy
reflects a desire to care for neglected things: “Caring is connected with aware-
ness of oppression, and with commitments to neglected experiences that
create oppositional standpoints” (Martin, Myers, & Viseu, 2015; Puig de la
Bellacasa, 2011, p. 96). Diversity advocates are right to explicate differential
burdens of vulnerability within technical cultures, as this analysis is essential
for a project of justice (Dunbar-Hester, 2020; Harding, 2016). Even if they
cannot fully ameliorate or redistribute these burdens, to care and to cultivate
seem to be the most important goals of diversity advocates (which are most
generative when not overly muddled with market logics). To maximize the
potential for generative justice (Eglash, 2016), diversity advocates can be on
the lookout for connections between their inspired homegrown efforts and
wider social movements and policy changes, which can maximize their DIY
yearnings in productive concert with structural change. We should also recog-
nize the imperfect yet infinite potential of coalition and care across difference.
Notes
1. I do not mean to conflate different hacking/FLOSS projects and subcultures, but
for the purposes of this analysis, they are grouped and considered more similar than
distinct. A different analysis might bring to bear significant features in order to draw
out contrast between sites. A one- or two-site study would provide more intimate
portraiture. Because I am interested in a larger perspective, this trade-off is acceptable.
2. Online discussion, Debian community, 2009 (https://debian-administration.org/
users/dkg/weblog/54).
3. I use the label “Hispanic” when referencing reports which have used that label.
4. Benjamin (2016) underscores that technoscientific innovation is a site for making and
remaking race and racism. Facial recognition software is inherently subject to racial-
ized abuses, even if it can be made more adept at recognizing a wider range of faces.
References
Ahmed, S. (2012). On being included. Durham, NC: Duke University Press.
Akrich, M. (1995). User representations: Practices, methods and sociology. In A. Rip, T.
J. Misa, & J. Schot (Eds.), Managing technology in society: The approach of constructive
technology assessment (pp. 167–184). London, UK: Pinter Publishers.
Banet-Weiser, S., & Miltner, K. M. (2016). MasculinitySoFragile: Culture, structure, and
networked misogyny. Feminist Media Studies, 16, 171–174.
Benjamin, R. (2016). Innovating inequity: If race is a technology, postracialism is the
Genius Bar. Ethnic and Racial Studies, 39, 2227–2234.
Callahan, B., Hathaway, C., & Krishnamoorthy, M. (2016). Quantitative metrics for gen-
erative justice: Graphing the value of diversity. Teknokultura, 13, 567–586.
Coleman, E. G. (2012). Coding freedom: The ethics and aesthetics of hacking. Princeton,
NJ: Princeton University Press.
Collins, P. H. (2000). Black feminist thought: Knowledge, consciousness, and the politics of
empowerment. New York, NY: Routledge.
Daniels, J. (2016). The trouble with white feminism: Whiteness, digital feminism and the
intersectional internet. In S. Noble & B. M. Tynes (Eds.), The intersectional inter-
net: Race, sex, class, and culture online (pp. 41–60). New York, NY: Peter Lang.
De la Peña, C. (2010). The history of technology, the resistance of archives, and the white-
ness of race. Technology and Culture, 51, 919–937.
Dunbar-Hester, C. (2016). Geek. In B. Peters (Ed.), Digital keywords (pp. 149–157).

Princeton, NJ: Princeton University Press.
Dunbar-Hester, C. (2019). If ‘diversity’ is the answer, what is the question? Understanding
diversity advocacy in open technology projects. In J. Vertesi & D. Ribes (Eds.),
digitalSTS: A field guide for science & technology studies (pp. 81–98). Princeton,
NJ: Princeton University Press.
Dunbar-Hester, C. (2020). Hacking diversity: Open technology cultures and the politics of
inclusion. Princeton, NJ: Princeton University Press.
Eglash, R. (2002). Race, sex, and nerds: From Black geeks to Asian American hipsters.
Social Text, 71(2), 49–64.
Eglash, R. (2016). An introduction to generative justice. Teknokultura, 13, 369–404.
Fouché, R. (2003). Black inventors in the age of segregation. Baltimore, MD: Johns
Hopkins University Press.
Fox, S., Ulgado, R., & Rosner, D. (2015). Hacking culture, not devices: Access and recog-
nition in feminist hackerspaces. In Proceedings of the 18th ACM conference on computer
supported cooperative work & social computing (pp. 56–68). Vancouver, BC: ACM.
Free Software Foundation. (2012, October 16). Happy Ada Lovelace Day! Retrieved from
https://www.fsf.org/blogs/community/happy-ada-lovelace-day
Geertz, C. (1998, October 22). Deep hanging out. New York Review of Books. Retrieved
from https://www.nybooks.com/articles/1998/10/22/deep-hanging-out/
Grenzfurthner, J., & Schneider, F. A. (2009). Hacking the spaces. Retrieved from http://
www.monochrom.at/hacking-the-spaces/
Harding, S. (1995). Just add women and stir? In Missing links: Gender equity in science and
technology for development, [compiled by] Gender Working Group, United Nations.
Commission on Science and Technology for Development (pp. 295–307). Ottawa,
Canada; International Development Research Centre.
Harding, S. (2016). Objectivity and diversity: Another logic of scientific research. Chicago,
IL: University of Chicago Press.
Jordan, T. (2017). A genealogy of hacking. Convergence, 23, 528–544.
Kearney, M. C. (2010). Pink technology: Mediamaking gear for girls. Camera
Obscura: Feminism, Culture, and Media Studies, 25, 2 (74), 1–39.
Kelty, C. (2005). Geeks, social imaginaries, and recursive publics. Cultural Anthropology,
20, 185–214.
Kelty, C. (2008). Two bits: The cultural significance of free software. Durham, NC: Duke
University Press.
Kreiss, D., Finn, M., & Turner, F. (2011). The limits of peer production: Some reminders
from Max Weber for the network society. New Media & Society, 13, 243–259.
Margolis, J. (2010). Stuck in the shallow end: Education, race, and computing. Cambridge,
MA: MIT Press.
Martin, A., Myers, N., & Viseu, A. (2015). The politics of care in technoscience. Social
Studies of Science, 45, 625–641.
Nafus, D. (2012). ‘Patches don’t have gender’: What is not open in open source soft-
ware. New Media & Society, 14, 669–683.
Nafus, D., Leach, J., & Krieger, B. (2006). Free/libre and open source software: Policy support
(FLOSSPOLS), Gender: Integrated Report of Findings. Cambridge, UK: University
of Cambridge. http://flosspols.merit.unu.edu/deliverables/FLOSSPOLS-D16-
Gender_Integrated_Report_of_Findings.pdf
Nakamura, L. (2013, December 10). Glitch racism. Culture Digitally (blog). Retrieved
from http://culturedigitally.org/2013/12/glitch-racism-networks-as-actors-within-
vernacular-internet-theory/
National Center for Women & Information Technology. (n.d.). https://www.ncwit.org.
Noble, S., & Tynes, B. (2016). The intersectional internet: Race, sex, class, and culture
online. New York, NY: Peter Lang.
Oudshoorn, N., Rommes, E., & Stienstra, M. (2004). Configuring the user as every-
body: Gender and design cultures in information and communication technologies.
Science, Technology & Human Values, 29, 30–63.
Polletta, F. (2002). Freedom is an endless meeting: Democracy in American social move-
ments. Chicago, IL: University of Chicago Press.
Poster, W. (2008). Filtering diversity: A global corporation struggles with race, class, and
gender in employment policy. American Behavioral Scientist, 52, 307–341.
Powell, A. (2012). Democratizing production through open source knowledge: From
open software to open hardware. Media, Culture & Society, 34, 691–708.
Puig de la Bellacasa, M. (2011). Matters of care in technoscience: Assembling neglected
things. Social Studies of Science, 41, 85–106.
Reagle, J. (2013, January 7). ‘Free as in sexist?’: Free culture and the gender gap. First
Monday, 18(1). https://firstmonday.org/article/view/4291/3381
Slaton, A. (2017, November 28). Exit, stage left: Towards transformative cri-
tiques of diversity. November 28. Retrieved from https://amyeslaton.com/
exit-stage-left-towards-transformative-critiques-of-diversity/
Streeter, T. (2010). The net effect: Technology, romanticism, capitalism. New York: New York
University Press.
Turner, F. (2009). Burning Man at Google: A cultural infrastructure for new media pro-
duction. New Media & Society, 11, 73–94.
8.
Religious Influencers: Faith in the
World of Marketing
Mara Einstein
The intersection of religion and marketing has garnered increasing attention

from religion, media, and marketing scholars and practitioners over the last
two decades (Atkin, 2004; Beaudoin, 2003; Belk, Wallendorf, & Sherry,
1989; Carrette & King, 2005; Cimino & Lattin, 1998; Einstein, 2008; Lee
& Sinitiere, 2009; Moore, 1994; Twitchell, 2004). Several factors have con-
tributed to this. First, religion and marketing share the idea of evangelism;
one promotes faith whereas the other works to engender enthusiasm for a
whole host of products and services. Second, because religion is no longer
ascribed but obtained, that is chosen, there is a robust marketplace of faith
in a way there has not been before. Further, the competition among belief
systems has intensified with the expansion of digital media platforms that have
allowed people to explore different and even multiple belief systems at the
push of a button in the privacy of their own home (McClure, 2017). Finally,
the rise of spirituality over traditional religion has provided the opportunity
for a wealth of belief-based products and practices to enter the marketplace.
This includes everything from designer prayer beads to crystals plugged by
Gwyneth Paltrow, from Marianne Williamson meditation tapes to TD Jakes
conferences to high-end spiritual travel to places such as Machu Picchu.
I have long argued that religion is a product (Einstein, 2008). As such,
it can be, and has been, promoted like other marketable goods. In the past,
this usually meant using television commercials such as the Rethink Church
campaign from the Methodist Church (United Methodist Communications,
n.d.) or Scientology’s annual Super Bowl commercial, which in 2019 was
entitled Curious? (Frazier, 2019). Print ads in newspapers alert people to
local church services. Websites present Sunday morning services to followers
around the world as was done with radio and TV in the past, but now these
122 Mara Einstein
include virtual pews available for real-time participant commentary. These

online services allow religion shoppers to sample what a congregation has
to offer before committing to it. In addition, denominations including the
Episcopal Church have also used their websites to share videos promoting the
variety of their churches—storefronts to cathedrals—as well as highlighting a
“legacy of inclusion” (Episcopal church, n.d.).
But these marketing tools are no longer as effective as they used to
be for religion or any other product, especially when communicating with
younger generations. Legacy media have given way not only to digital plat-
forms but also to social media outlets from Facebook to Twitter to YouTube
and SnapChat. Importantly, in these social spaces, visitors do not want their
online experience interrupted with advertising. In response, marketers found
ways to make commercial messages less obvious—first with native advertising
(ads made to appear indigenous to the site in which they appear) and then
with branded content.
Over the last few years, branded content has evolved across all media
forms. Branded content, also known as branded journalism, brand storytell-
ing, and most broadly as stealth marketing, is defined as content provided by
marketers that is deemed valuable by consumers yet does not contain an overt
sales message (Einstein, 2016). From the New York Times’ Brand Studio to
CNN’s Courageous Studio to BuzzFeed’s listicles and cat memes, advertisers
are producing information and entertainment that disguise the commercial
interest behind it. Most people do not understand that they are engaging
with information that has a biased producer behind it. Research has shown
that only 17% percent of people recognize native advertising, and even digital
natives have a difficult time discerning when they are engaging with advertis-
ing rather than editorial content (Amazeen & Wojdynski, 2018; Wineburg,
McGrew, Breakstone, & Ortega, 2016; Wojdynski & Evans, 2015).
The natural evolution of covert marketing has been the widespread use
of influencers to sell products without acknowledging they are being paid for
endorsing a product, an inevitable extension of the branded content strategy.
The Internet Advertising Bureau (IAB), the industry’s leading digital adver-
tising association, defines influencers as “those who are deemed to have the
potential to create engagement, drive conversation and/or sell products/ser-
vices with the intended target audience. These individuals can range from being
celebrities to more micro-targeted professional or nonprofessional ‘peers’ ”
(IAB, 2018). You might say, “So what?” and that consumers know when they
are looking at an influencer message. They know when Kim Kardashian is selling
a vitamin, or Huda Kattan is selling the latest beauty product. Well, yes and no.
In the United States, people who are paid to promote a product are required to
Religious Influencers 123
disclose that on their posts by using #ad or #sponsored, for example. However,
most influencers are not in compliance with these requirements, and even after
being called out by the Federal Trade Commission (FTC) for non-compliance,
influencers continue to flout the system (Public Citizen, 2017; Maheshwari,
2016). Moreover, as influencer marketing took off, so did the price tag for
paying influencers. Marketers began to look for inexpensive ways to employ
this tactic. One strategy is to hire teenagers who will post an endorsement for as
little as $10, with no indication of its sponsor messaged status (Lorenz, 2018).
So although visitors might suspect a celebrity is shilling for a product, they are
less likely to think the same thing about an unknown teenager. Even celebrities
may not be on the up-and-up when it comes to this type of promotion. The
best example of this is the massively hyped Fyre Festival, an influencer-fueled
snafu of an event that never materialized. Fyre, a Netflix documentary about
the event highlighting the schemes used to bamboozle wealthy, influencer-fol-
lowing people into spending thousands of dollars, was produced by one of the
promotional companies perpetrating the fraud. The company made the film to
help whitewash its image while never acknowledging any connection to the film
(Livingstone, 2019). Talk about branded content! Even with all this, influenc-
ers are now the go-to form of advertising.
In this chapter, I examine the rise of influencer marketing and how
social media influencers are segmented both according to product verticals—
beauty, parenting, fashion, etc.—and by audience size. I will then examine
how this marketing strategy is being utilized within the religion category.
Unlike beauty products, however, the product being sold can often be amor-
phous. Is it the Bible or a church leader or Christianity or even clothes with
a religious twist that is being hawked? Understanding religious influencers
turned out to be a much larger and more complex landscape than I thought.
To better grasp its many levels, I have conceptualized a taxonomy of religious
influencers, which I present below.
Branded Content and the Rise of Social Influencers

As social media rose in popularity, publishers and marketers were most con-
cerned about how to churn out content—the more content a company
produced, the higher up on search engines one might appear. Content com-
panies, such as NewsCred and Contently, arose and maintained stables of
thousands of freelancers—many of whom were former journalists—available
to write stories at a moment’s notice. Simultaneously, technology for selling
advertising expanded so that most ad buying became what is called “program-
matic,” or done by computers. Increasing amounts of new content married
124 Mara Einstein
with persuasive design technology led to more time spent online, which led to
more consumer data for marketers and publishers to analyze. Thus, creating
content is easier, selling ads is faster, people are spending more time online,
and tracking consumers has become increasingly automated and less labor
intensive. For marketers and even more for platforms such as Facebook, it is
a win-win-win. For consumers, not so much. Few know or understand the
depth to which their data is collected to customize their advertising experi-
ence, how little it is anonymized, or how it is being used against their best
interests (Nielsen, 2018; Valentino-DeVries, Singer, Keller, & Krolik, 2018).
Nor do most grasp the extent to which marketers want to integrate what they
do online with offline buying and sharing behaviors.
The integration of online and offline marketing components became par-
amount for marketers, and we saw this in the way that campaigns combined
these elements. For example, consider Fearless Girl, a famous Wall Street
statue created by a financial firm, State Street Global Advisors, to bring atten-
tion to the lack of women on corporate boards and to promote an investment
opportunity: a fund that invests in companies the support more women in
leadership positions. This unassuming statue—that is, an ad—has become a
New York City destination which gets promoted over and over, thanks to the
thousands of tourists who pose with the statue and post selfies to social media.
(We could have a discussion about the use of a girl to represent women’s
empowerment or the hypocrisy that State Street was forced to pay $5 million
to settle a lawsuit for gender pay discrimination (Bellstrom, 2017).) In a more
elaborate example, Lady Gaga worked with Intel to create a tribute to David
Bowie during a Grammys telecast. Viewers witnessed the seamless integration
of music, dance, and technology. They also posted to social media where they
were presented with more information about the technology as well as videos
of Lady Gaga touting the wonders of Intel (Insomnia, 2017). Finally, one of
the most elaborate integrated campaigns I’ve found is one for MailChimp, a
popular email distribution platform. Playing off the idea that people continu-
ously get the brand name incorrect, advertising agency Droga5 created nine
fake brands with names such as Kale Limp and Male Crimp, and produced
several marketing elements including ads, films to show online and at movie
theaters, websites, Tumblrs, and even a hit song called Veil Hymn. If someone
searched for one of these on Google, the site asked if you meant to look for
MailChimp, which is an incredibly smart search engine optimization (SEO)
strategy. In sum, then, what we have seen in the evolution of branded content
is a continued increase in video over static creative, more complex executions
that combine online and offline elements, and an increasingly seamless inte-
gration of editorial and advertising content.
The consumer journey online has also changed. Traditionally, marketers

have thought about the transition from unaware shopper to product pur-
chaser as that of a funnel. At the top of the funnel, marketers are looking
to create awareness, and along the way, consumers become more and more
invested in the brand the more they interact with it. In the past, creating
awareness meant television commercials, and consumer engagement might
mean a trip to the local retail outlet culminating in a purchase. Today, con-
sumers are consistently presented with brand messages, so the idea of a sales
funnel no longer makes sense. Instead, the consumer decision process is now
most often talked about as a loyalty loop where consumers and marketers
are in a continuous loop of engagement (Court, Elzinga, & Mulder, n.d.).
Some practitioners have gone so far as to perceive the marketing process as
collapsed into a single step; our mobile devices allow us to see, research,
and purchase a product all in the same place. Mary Meeker’s annual Internet
Trends Report is one of the most widely read sources of information about the
topic, and in 2017 she noted, “The content is becoming the store … the ad
is becoming the transaction” (Meeker, 2017). Now I would argue the influ-
encer has become the content and the ad.
Types of Influencers
Although content is important, the purveyor of that content is paramount
for marketers. Online interactions are more personal and demand more emo-
tional connection (Berger, 2013) than legacy media. Our phones are per-
sonal, so much so that they are extensions of ourselves, and our social media
interactions are one-on-one, or at least they feel that way.
Because of this, influencers provide significant advantages for digital
marketers. First, influencers have created relationships—seemingly trusting,
authentic relationships—with their followers, which make them perfect brand
ambassadors, whether selling Raisin Bran or religion. Second, influencers are
content producers as well as message disseminators. The available level of
customization enables higher engagement levels than found in other digital
formats. Third, influencers offer a two-way conversation. Visitors/consum-
ers might interact with a brand or their church, but they are far more likely
to have a conversation with one of their favorite YouTubers or Instagram
celebrities. And, unlike most brands or religious institutions, influencers are
entertaining and relatable; they evoke emotion and have a relationship with
the visitor on a channel the audience opted into.
Typically, when we think of influencers, we think of people who have mil-
lions of followers across multiple online platforms. PewDiePie has more than
126 Mara Einstein
4 million subscribers on YouTube who watch his video game commentary;

he racks up another 16 million followers on Twitter and the same number on
Instagram. Jenna Marbles disseminates comedy to over 27 million followers
across YouTube, Instagram, Twitter, and Facebook. Influencers such as these,
as well as traditional celebrities such as Beyoncé (126 million on Instagram) or
Lady Gaga (78 million on Twitter), provide loyalty as well as media tonnage.
But not all influencers have millions of followers, nor do they have
to (Collectively, 2018). Followers as a measure of marketing success has
come into dispute with the rise of bots and the ability to buy followers (see
Confessore, Dance, Harris, & Hansen, 2018), yet it continues to be used as a
key performance indicator. The true value of influencers, however, is in their
relationships with their audiences and their ability to get those audiences to
buy into what they are selling, whether it is a product or themselves. Broadly,
then, there are thousands of influencers with follower numbers ranging from
a few hundred to many millions. At the top of the list are the mega influenc-
ers. At the next level are micro-influencers, with tens of thousands of follow-
ers, who can be important in persuading their fan base if their followers are
loyal and passionate. Finally, there are nano influencers, with a few thousand
followers, who present themselves as everyday people. These nano influenc-
ers have become increasing popular as star influencers have become more
expensive, and according to Maheshwari, “Brands enjoy working with them
partly because … in exchange for free products or a small commission, nanos
typically say whatever companies tell them to” (2018). Most advertising cam-
paigns use more than one influencer, with 73% of campaigns using more than
10 (Linqia, 2018). These handful or hundreds or even thousands of influ-
encers are provided via influencer networks, companies that maintain pools
of influencers—much like content farms maintain freelance writers—who are
matched up with marketers and marketing campaigns typically through an
online platform.
In the branding world, these influencers are segmented into verticals, or
product categories. Beauty, fitness, parenting, and pets are some of the most
popular verticals. Grouping influencers in this way makes it easier to connect
influencers with advertisers. And just like cars or crafting, religion is a market
segment, or vertical.
Religious Influencers
Religious influencers are unique. Unlike other product categories, faith is
not homogeneous. There are different religions, different forms of practice,
and a variety of ways to promote belief systems both within and outside of a
traditional religious institution. In addition, many influencers present them-

selves as religious to promote secular products rather than to promote reli-
gion per se.
Creating a typology for religious influencers was not as straightforward
as I suspected it would be. Having researched media and religion for more
than a decade, I predicted that religion would follow the patterns of other
consumer marketplaces (which it did) but would also bring to the fore influ-
encers that had found creative ways to use social media to attract followers,
and particularly young people, to faith (which it did, but not very much).
Religious influencers reflect their secular counterparts in that most are sell-
ing products. These products may have a religious or spiritual bent to them,
or the products may be utterly secular, but followers buy into the clothes
or stationery or new-fangled food because they relate to the influencer and
her/his professed religious orientation. Influencers from traditional religious
institutions, on the other hand, were more likely to promote religion itself. In
addition, I assumed a broad definition of faith, how or whether the influencer
was connected to an offline institution or was online only, and whether reli-
gion was being exploited within this consumer framework. Questions I had to
grapple with included: Are you a religious influencer if you write about going
to church and how God helped you in your life but all you sell are bathing
suits and cosmetics? If what you sell is specifically related to your work—a
book, a TV show, a podcast—is that a different type of religious influencer?
And, what if you didn’t actually sell anything? Are you still an influencer?
Thus, a religion/spiritual to consumerist orientation guided my think-
ing in creating this typology. I began by looking at religious figures with the
most significant followings, that is, mega influencers. I turned to social media
outlets—primarily Twitter and Instagram because those are the top sites for
influencers—and ran searches on religions and different faith groups as well as
researching press outlets and influencer networks. Perhaps not surprisingly, the
only network to have religious influencers was Amazon, and this has become
limited over time as other consumer product verticals drive more traffic.
Based on my analysis, religious influencers exist on a spectrum from pure
promotion of faith to using faith to promote secular products (Traditional
Religious Leaders, Religious Innovators, Religious Social Celebrities, Faith-
filled Celebrities & Industrialists, Evangelicals/Televangelists, and False
Profits). At the most faith-focused end are traditional religious leaders, such
as the Pope. These believers are not so much selling faith as they are living
their faith, and in presenting it affecting those who follow them. At the other
end of the spectrum are influencers who primarily sell a product but use faith
as a tool to persuade prospective customers. Influencers are, by definition,
128 Mara Einstein
trying to influence a purchase. Here, though, not every influencer has the
mission to increase a marketer’s bottom line—another unique characteristic
of these influencers. In addition, some categories follow the mega-micro-
nano framework of secular influencers and some do not. Finally, this typology
is by no means definitive but a work in progress meant to begin a dialogue
about religion and marketing in this new—and evolving—media space.
Traditional Religious Leaders

Influencers in this category have a following due to leading a traditional reli-
gious congregation. Followings in the physical world through a brick-and-
mortar church, synagogue, or mosque bolster the interaction with audiences,
that is, integrated online/offline communication. Like branded content,
these influencers do not have an overt sales message.
Social media is used by a wide swath of religious leaders connected to
religious institutions. By far, the two most popular figures in this category are
Pope Francis and the Dalai Lama. Pope Francis uses Twitter and Instagram,
a site where he amassed more than over 1 million followers within 12 hours
of creating an Instagram account. Both men have followers in the millions.
More importantly, they have significant engagement with their followers.
Pope Francis, who tweets regularly but infrequently (approximately 6 tweets
per week), generates on average more than 8500 retweets and more than
19,500 likes (O’Loughlin, 2017). The Dalai Lama has 19 million followers
on Twitter, and his tweets commonly receive tens of thousands and even
more than 100 thousand likes. Many of his tweets are short teachings on
compassion rather than overtly religious content. Here is an example of a
typical Dalai Lama post:
Of the 7 billion human beings alive today, no one wants to suffer; no one chooses
to have problems. Yet, many of the problems we face are our own creation. Why?
Because of ignorance. But ignorance is not permanent and whether we overcome
it depends on whether we make the effort.
The Pope’s most popular tweets are decidedly religious:

Holy Spirit, our harmony, You who make us one body, infuse your peace in the
Church and in the world!
Holy #MaryMotherOfTheChurch, help us to entrust ourselves fully to Jesus
and to believe in His love, especially in times of trial, beneath the shadow of the
Cross, when our faith is called to mature.
Although these religious leaders have a similar number of followers, the post
from the Dalai Lama received more than 87 thousand likes versus 30 thousand
for the Pope. Much like traditional media, a more generic religious message
has broader appeal and better engagement.
It is interesting, too, to note the difference in design of the social media
pages of these leaders versus influencers in other religious segments. Pope
Francis’s Twitter page, for example, has a picture of him smiling, standing
amid a group of people. The main image on the Dalai Lama’s page is a simple
burnt red bar similar to the color of the robe he wears. The simplicity of the
page forces attention to the messages being sent.
The strata exists in this segment of religious influencers. Micro influenc-
ers in this category are people such as PB Michael Curry, the presiding bishop
of the Episcopal Church (@PB_Curry) with 67 thousand followers on Twitter
and another 32 thousand on Instagram. Nano influencers in this category
include the rabbi, pastor, or imam of local congregations.
Religious Innovators
Religious innovators use social media as a new way to engage with an audi-
ence, either by providing fresh ways to explain religious traditions to current
practitioners or by introducing a faith to those not already affiliated.
When I began this research, I expected this group to be the bulk of what
I would find. I expected to see pastors teaching online or coming up with
new ways to present religious material. Although church services are offered
online as a matter of course similar to offering these services via radio or TV,
finding a religious leader doing something truly unique was harder.
Who I did find was Daniel Bortz, known as the Millennial Rabbi. True
to the best ways of using the medium, Rabbi Bortz is approachable and per-
sonable in his online presentation. On Instagram, we see him walking on
Santa Monica pier or standing in Grand Central Station or leaning up against
a wall covered with graffiti. He competes in Ju Jitsu and is the rabbi for the
Coachella Music Festival—a point of millennial cred if there ever was one.
The rabbi is active on multiple social media sites including Snapchat, but is
especially visible on Instagram, which makes sense given his young audience
and is typical for active influencers.
What makes him stand out, however, is how he uses these platforms to
teach his faith. He understands that the various social media platforms work
synergistically to communicate his message but that each does so in different
ways requiring different messaging. On Snapchat, he knows he can be less for-
mal and polished, so this is where he posts behind-the-scenes or light-hearted
content, which makes him seem real and approachable. On Facebook, the
rabbi posts videos where he talks directly to the camera. Most notable were
teachings he did on Instagram for Hanukkah. He created content for each of
130 Mara Einstein
the eight days of the holiday. As he explained it, “There were to be eight days
of carefully curated Insta stories, including a live stream ‘delving deeper’ into
the Jewish holiday, accompanied by a post quoting from Proverbs 20:27 and
exploring the symbolism of flames and natural light” (BBC, 2018).
This made me think of one of the best religion-related uses of Twitter
I have ever seen. In 2009, Trinity Church in lower Manhattan re-enacted the
Passion on Good Friday. Over the course of three hours, the church posted
tweets from Jesus, Mary, Pilate, Joseph, and Peter, describing the events lead-
ing up to Jesus being put on the cross (Lee, 2009). The brilliance of this from
a marketing perspective is that the communications were reaching a young
audience that was attending church less often than in the past. At the same
time, the church was communicating to its audience members where they
were and for a continuous and extended period of time. This was a tweet
thread before the term even existed, and it presented an old story in a brand
new way, much as what Rabbi Bortz did with Hanukkah.
Rachel Held Evans was another religious innovator, who died at the age
of 37 in May 2019. As her New York Times obituary noted: “Her congrega-
tion was online, and her Twitter feed became her church, a gathering place
for thousands to question, find safety in their doubts and learn to believe in
new ways” (Dias & Roberts, 2019).
Evans is best known for having left evangelicalism because of its con-
straints and exclusivity. One of her tweets noted the reason, “When I left
church at age 29, full of doubt and disillusionment I wasn’t looking for a
better-produced Christianity. I was looking for a truer Christianity, a more
authentic Christianity.” Her own experiences led her to write books about
leaving a faith she found so confining. Although the books introduced her to
audiences, it was her social media presence that enabled the diaspora of ostra-
cized groups to find a place to question church practices. Reading Rachel’s
tweets questioning faith opened the door for others to do the same. Debating
and questioning religion can be difficult under the best of circumstance. This
Twitter feed provided the space for that to happen, something that works on
social media because of its feeling of being one-on-one and the accessibility
of the influencer. At the time of her death, Rachel Held Evans had 170 thou-
sand followers.
Finally, @preachersnsneakers is an Instagram page that posts pictures of
popular megachurch religious figures wearing high-end designer clothing and
shoes, particularly sneakers. Alongside the pictures of the pastors are pictures
of the sneakers or track pants or Gucci backpack with the price of the item
prominently displayed. The anonymous poster—an avowed sneakerhead—
began the page because he noticed $900+ shoes on the feet of people who
at least on the face of it promoted a prosperity gospel that serves them more
than their followers. Although not preaching the faith directly, the site has
opened up an important ethical conversation (Christian, 2019).
Religious Social Celebrities

Like traditional influencers, these online celebrities have found fame on social
media because of their religion. Unlike leaders of religious institutions, these
are laypeople and although they are upfront about their faith they also tend
to provide commentary related to another vertical; they are religious, but
they are also mommy bloggers or travel bloggers. Because they practice at
the intersection of two verticals, they tend to be micro influencers rather than
mega ones.
Chaviva Galatz began her career as a blogger, tweeter, and social media
personality. Her claim to fame was that she created and co-chaired the only
Jewish-themed panel at the 2011 South by Southwest (SxSW) Interactive
Festival: Jewish Synergy: Social Media and the New Community. Blogging
under the title Just Call Me Chaviva, Galatz chronicled her story about con-
verting to Judaism. More recently, Galatz has combined her faith stories with
posts about food and her kids. She has over time morphed into a mommy
blogger, one of the most popular influencer segments. She promotes food
stores, recipes, and books. There is no indication that she is being paid for
this, though given her following and her open manner, she easily could be.
Elena Nikolova’s story is a bit different because her faith both informs
and is integral to her online presence. Nikolova founded Muslim Travel Girl
in 2013 to help Western Muslims with Halal travel. After discovering the
benefits of airline loyalty programs, her friends asked for assistance on how to
travel business class without the high fees; what started as a simple blog about
loyalty programs became a leading blog for Muslim-friendly travel. Given
the category, it is no surprise that Instagram is the primary social media used
(44.3K followers). Elena posts beautiful pictures of exotic destinations (often
with her in her hijab) while peppering her posts with religious well-wish-
ing (“This Ramadan flew by but I pray our sins are forgiven we have our
duas answered and we enjoy the next 11 months in good health until next
Ramadan.”) I put her in this category rather than Religious Marketers because
of what she says on her blog: “I am on a quest to improve mutual understand-
ing by helping Muslims travel the world” (Muslim Travel Girl, n.d.)
Faith-filled Celebrities and Industrialists

Celebrities and successful businesspeople carry considerable weight on social
media much as they do in traditional media. Advertising has long tapped the
talent of celebrities to promote products. From Michael Jordan and Nike to
132 Mara Einstein
Katy Perry and Cover Girl, using a celebrity to attract attention to a prod-
uct generates what is known as borrowed interest. That is, the advertising
depends on the fact that people will be interested in the celebrity and will
attend to the marketer’s message. In the same vein, we see actors or music
artists thank God or Jesus when they accept an industry award—they know
audiences care about their movie or music and use their time in the spotlight
to bring attention to their religion. Few celebrities promote their faith much
further than that. However, a handful of celebrities and businesspeople con-
sistently connect their communications to religious tenets.
Here I present two extreme examples for demonstration’s sake. Tim
Tebow, a former NFL quarterback who left football for baseball, became
famous not only for his skills as an athlete but also for kneeling to pray both
before and during football games. Dropping to his knee to with his head
bowed in prayer became so well known that it was dubbed “Tebowing”
(Frost, 2017). From the beginning of his career, he was open about being
homeschooled, pro-life, and devoutly, conservatively Christian. Tebow’s
professions of faith have been translated to social media and are mixed in
with sports-related content. On Instagram, he has 1.6 million followers and
another 4.8 million on Twitter. Unlike the other groups, we’ve looked at thus
far, his social media use surreptitiously promotes teachings from his evan-
gelical roots. In early 2019, for example, he promoted a movie called Run
the Race, which at first glance appears to be a typical Hollywood release.
However, the movie’s website (runtheracemovie.com) reveals otherwise and
provides resources supporting Bible study to be used in conjunction with the
film. The film has partnered with Christian Healthcare Ministries, All Pro
Dad, and National Coalition of Ministries to Men.
Note that using social media in this way parallels strategies used on legacy
media. When communicating with large groups of people, overt religious
content can be polarizing. Therefore, mediated religion was walled off on
spaces where people specifically opted in to see it. In one example, Oprah did
an online series with Eckhart Tolle about spirituality that required people to
opt in. Another strategy is to separate religious content from the main con-
tent being presented so that viewers could easily leave when the religious con-
tent got too overt, a strategy often used by televangelists (Einstein, 2008).
Evangelicals/Televangelists
These congregational leaders sell their books and their politics as much
as they sell their belief system, which may be why they have been dubbed
“entrepreneurial evangelicals” (Burke, 2016). Note that these are not just
Christians; Rabbi Shmuley, famous for being Michael Jackson’s and Roseanne
Barr’s spiritual adviser and for writing dozens of books, including Kosher Sex,
is included in this group.
Evangelicals—or simply religious marketers—thrive in the world of social
media. What they do is all about selling people on the word of God, although
most of them are doing a knockout job of promoting themselves at the same
time. T.D. Jakes, Joel Osteen, Rick Warren, Paula White, and Joyce Meyer are
just a handful of the people representing this category. Like the Traditional
Religious Leaders, most are connected to an offline church. However, evan-
gelicals head megachurches, serving tens of thousands in buildings that are
more shopping mall than house of worship. Most—though not all, notably
Rick Warren—are also televangelists who have honed their communications
skills overs years on broadcast or cable TV. The combination of these fol-
lowings online and offline allows them to take advantage of the interplay of
multiple forms of engagement that feed on one another.
Evangelicalism is, by definition, about selling the faith, and this group
more than any other is about promotion, promotion, promotion. They pro-
duce more content and generate significantly more engagements and retweets
than other groups, so much so that they even rival rock stars. Their tweets
get likes in the tens of thousands, and their Instagram posts get over 100,000
likes. And, because these influencers produce for other media, they have a
continuous cycle of new content that can be repurposed for social media plat-
forms. So Joel Osteen or Bishop Jakes can take a small section of his weekly
sermon and post it on Instagram with titles over what is being said so that
people can watch the video without sound. These snippets get tens of thou-
sands of views.
Another example is Christine Caine, who like others in this category uses
social media to sell books and promote her appearances and workshops. Yet
unlike the other televangelists, Caine promotes an empowering message spe-
cifically targeted at women. This is tied to her work with her organization,
Propel Women, which works to support women in the Christian faith, includ-
ing women in ministry. In addition to her preaching, Caine and her husband
founded A21, an organization that works to end slavery and sex trafficking
around the world. Posts related to this work fill the feed, alongside scripture
and words of empowerment.
False Profits
False profits are salespeople who sell secular or sometimes spiritually based
products while infusing religion—or the appearance of being religious—in
their communications. These influencers are selling everything from clothes
to prayer books, but it is their claims to faith that endear their followers
134 Mara Einstein
to them. They are most like secular influencers, selling their products to a
defined target audience. Portraying oneself as religious or spiritual engenders
trust in a way that almost nothing else can.
Ashley Brown (@ashleyempowers) posts about relationships and religion.
Her website looks like a less-impressive version of an evangelical pastor’s site.
She promotes herself as “a speaker, entrepreneur, author and top media per-
sonality” and the photography on the site (and on social media) is obviously
professionally done. One of the tabs on her AshleyEmpowers.com site is
called “Relationship Empowerment,” and outlines various learning modules
of a course called “MAN-ifest Your Godly Man” and ends with to the cost
of taking the course—as of this writing $99, on sale from $199. Brown is
even more effective on social media. She has more than 66,000 followers
on YouTube and 25,000 followers on Instagram, making her a niche influ-
encer. Her videos have titles such as, “God told me who my husband is,”
“Why I don’t take birth control,” and “How should I respond to an unwed
mother?” Her followers respond with messages such as, “Thank you for shar-
ing Ashley! I needed this. God bless you and your family!” and “God has used
you to speak to me xx.” On Instagram, she is both personal (she posts about
trying to lose weight after having a second child) and more promotional than
on YouTube. Most of her posts include the hashtag #ad, and she sells every-
thing from Walmart to Cheerios to Carol’s Daughter hair products and @fash-
ionnova. She also promotes charitable organizations such as Stand Together,
an organization that matches volunteers to non-profit organizations.
Another example is Sazan Hendrix, a fashion and beauty blogger, and
co-founder of Bless Box. Similar to other subscription boxes, Bless Box
contains beauty, health, and food products curated for the subscriber. The
“bless” moniker is tenuous in its relationship to religion, much as SoulCycle
has little to do with God or goodness. On the Bless Blog, for example, featured
stories have titles such as “The Bless Guide to Saving a Manicure from Chips,
Splits & Cracks” “The Bless Guide to Beating Scalp Build-Up (& Getting
Gorgeous Hair),” and “The Essential White Sneakers to Wear this Spring,”
which are not topics one would read about in the Bible or Quran. Hendrix
is a media powerhouse with a website (Sazan.me), nearly 12,000 followers
on Twitter (which she primarily uses to direct people to her Instagram where
she has one million followers) and a YouTube channel with over 400 thou-
sand subscribers. A religious vibe is subtly suffused through her work. On
Instagram, for example, she posted a picture of her daughter with the caption,
“I can’t help but look at this sweet angel and think wow God is SO good.”
On her blog, she posted about her collaboration with Cindy Crawford say-
ing, “I thank God that I found the strength to get ready and go to Cindy’s
Meaningful Beauty event because she was exactly what I needed. I’ll never
forget all the advice and wisdom she shared with me that night that blessed
me so much.” (She also stated, “I was never planning to post the footage
I recorded from that night,” which is hard to believe when you look at the
high production value of the content.) On top of all of this, she produces a
podcast with her husband called The Good Life with Stevie & Sazan where
they talk about family and fashion. Advertising—overt or hidden—pervades
her platforms. Listeners complain about the amount of advertising in this
podcast, and her Instagram is filled with posts about beauty products without
disclosing whether she was paid for her endorsement.
Religion, Social Media, and the Market

Religion provides an opportunity to conceive of influencer marketing more
beyond a capital market context. In doing so, we can begin to think of influ-
encers as defining success through a broader range of metrics than simply
product sales or revenues generated. Influencers have the ability to impact
our thinking about politics, education, and, yes, religion, to name a few. What
makes this more difficult to assess is that these social and cultural institutions
are more complex, and ideas are more nuanced than “buy this product.” As
we have seen with religion, the category is not homogeneous and the goals
of the influencers are different—some do want sales; but others want to help
people, or bring people to God, or help people question their faith.
On one end of the religious influencer spectrum are religious leaders, for
whom social media becomes a means to communicate with younger followers
or prospective followers, a way to reach them where they are. The product
being sold is the religion itself and the tenets of the faith. Importantly, it is
not just the number of followers someone has but how much those followers
trust and believe in the messenger and whether that trust will translate into
action. Whether it is because of longevity, the backing of an institution, or the
force of personality, traditional religious leaders can get followers to engage
with them in numbers not seen in the other categories. Indeed, an under-
standing of the importance and personal nature of religion is what fuels the
many influencers using faith as part of their toolkit. For some, such as Rabbi
Bortz or Chaviva Galatz, faith is integral to their online communication. As
we move further along the spectrum, we see an increased market orientation
on the influencers—so much so that by the time we reach the False Profits,
there is very little to distinguish these social media celebrities from secular
ones, except for a mention of God here and a tip of the hat to Jesus there.
Let me be clear, I don’t know whether the people I use as examples here
136 Mara Einstein
are religious or not, and that is not the point. What I am concerned about
is that other influencers will try to use a religious guise to gain audience and
engender trust—sort of the prosperity Gospel of social media. This has conse-
quences beyond buying a dress or believing that God will bring your husband
to your door. It has the potential to become an exploitive tactic of persuasion,
harming not only the perceptions of religions and their leaders but also the
followers who may begin to see themselves as consumers rather than seekers
or congregants. Given the market connotations of influencer, it might be wise
to consider an alternative term for those who wish to promote in faith.
References
Amazeen, M. A., & Wojdynski, B. W. (2018). Reducing native advertising decep-
tion: Revisiting the antecedents and consequences of persuasion knowledge in digital
news contexts. Mass Communication and Society, 22(2), 222–247.
Atkin, D. (2004). The culting of brands: Turn your customers into true believers. New York,
NY: Portfolio.
BBC. (2018, March 7). Meet the religious social media stars: Their Instagrams are all about
god and spirituality. BBC.com. Retrieved from https://www.bbc.co.uk/bbcthree/
article/e940a06d-ea0a-4f5d-b8ac-6b3111934f9d
Beaudoin, T. (2003). Consuming faith: Integrating who we are with what we buy. Chicago,
IL: Sheed & Ward.
Belk, R. W., Wallendorf, M., & Sherry, J. F. (1989). The sacred and the profane in con-
sumer behavior: Theodicy on the odyssey. Journal of Consumer Research, 16(1), 1–38.
Bellstrom, K. (2017, October 5). The firm behind ‘Fearless Girl’ agreed to a $5 million
settlement—Over gender pay discrimination. Fortune. Retrieved from http://for-
tune.com/2017/10/05/fearless-girl-state-street-pay-discrimination/
Berger, J. (2013). Contagious: Why things catch on (1st Simon & Schuster hardcover ed.).
New York, NY: Simon & Schuster.
Burke, D. (2016). 7 types of evangelicals—And how they’ll affect the presidential race.
CNN.com. Retrieved from https://www.cnn.com/2016/01/22/politics/sev-
en-types-of-evangelicals-and-the-primaries/index.html
Carrette, J., & King, R. (2005). Selling spirituality: The silent takeover of religion. London,
UK: Routledge.
Christian, S. (2019, April 12). WWJD: The ethical conundrum of mega-church preachers
in super-expensive sneakers. Esquire.com. Retrieved from https://www.esquire.com/
style/mens-fashion/a27114633/expensive-sneakers-preachers-sneakersnpreachers-
instagram-interview/
Cimino, R., & Lattin, D. (1998). Shopping for faith: American religion in the new millen-
nium. San Francisco, CA: Jossey-Bass.
Collectively. (2018). The social influence business in 2018 & beyond. Retrieved from
https://collectivelyinc.com/reports.
Confessore, N., Dance, G., Harris, R., & Hansen, M. (2018, January 27). The Follower
Factory. The New York Times. Retrieved from https://www.nytimes.com/interac-
tive/2018/01/27/technology/social-media-bots.html
Court, D., Elzinga, D., & Mulder, S. (n.d.). The consumer decision journey. McKinsey.
com. Retrieved from https://www.mckinsey.com/business-functions/
marketing-and-sales/our-insights/the-consumer-decision-journey
Dias, E., & Roberts, S. (2019, May 4). Rachel Held Evans, voice of the wandering
evangelical, dies at 37. The New York Times. Retrieved from https://www.nytimes.
com/2019/05/04/us/rachel-held-evans.html
Einstein, M. (2008). Brands of faith: Marketing religion in a commercial age. London,
UK: Routledge.
Einstein, M. (2016). Black ops advertising: Native ads, content marketing and the covert
world of the digital sell. New York, NY: OR Books.
Episcopal Church. (n.d.). What we believe. Retrieved from https://www.episcopalchurch.
org/what-we-believe
Frazier, K. (2019, February 4). Scientology’s 2019 Super Bowl ad asks “Curious?”
World Religion News. Retrieved from https://www.worldreligionnews.com/
entertainment/scientologys-2019-super-bowl-ad-asks-curious
Frost, M. (2017). Colin Kaepernick vs. Tim Tebow: Christianity on its knees. Washington
Post. Retrieved from https://www.washingtonpost.com/news/acts-of-faith/
wp/2017/09/24/colin-kaepernick-vs-tim-tebow-a-tale-of-two-christianities-on-its-
knees/?utm_term=.0a8f4be724a4
IAB. (2018). IAB influencer marketing for publishers. Retrieved from https://www.
iab.com/wp-content/uploads/2018/01/IAB_Influencer_Marketing_for_
Publishers_2018-01-25.pdf
Insomnia. (2017, July 24). Intel and Lady Gaga at the Grammy Awards 2017. Retrieved
from https://www.youtube.com/watch?v=3DWie9JRmjo
Lee, J. (2009, April 10). Jesus tweets for a Good Friday. The New York Times.
Retrieved from https://cityroom.blogs.nytimes.com/2009/04/10/
jesus-tweets-for-a-good-friday/
Lee, S., & Sinitiere, P. (2009). Holy mavericks: Evangelical innovators and the spiritual
marketplace. New York, NY: NYU Press.
Linqia. (2018). State of influencer marketing. 2018. Retrieved from https://linqia.com/
wp-content/uploads/2017/12/Linqia-The-State-of-Influencer-Marketing-2018.
pdf.
Livingstone, J. (2019, February 12). Fyre Festival was a huge scam. Is Netflix’s Fyre doc-
umentary a scam, too? The New Republic. Retrieved from https://newrepublic.com/
article/153095/fyre-festival-huge-scam-netflixs-fyre-documentary-scam-too
Lorenz, T. (2018, August 22). Posting Instagram sponsored content is the new
summer job. The Atlantic. Retrieved from https://www.theatlantic.com/
technology/archive/2018/08/posting-instagram-sponsored-content-is-the-
new-summer-job/568108/
138 Mara Einstein
Maheshwari, S. (2016). Endorsed on Instagram by a Kardashian, but is it love or just

an ad? The New York Times. https://www.nytimes.com/2016/08/30/business/
media/instagram-ads-marketing-kardashian.html
Maheshwari, S. (2018). Are you ready for the nanoinfluencers? The New York Times.
Retrieved from https://www.nytimes.com/2018/11/11/business/media/nanoin-
fluencers-instagram-influencers.html
McClure, P. (2017). Tinkering with technology and religion in the digital age: The effects
of internet use on religious belief, behavior, and belonging. Journal for the Scientific
Study of Religion, 56(3), 481–497.
Meeker, M. (2017). Internet trends 2017—Code conference. Presentation on behalf of
Kleiner Perkins. May 31, 2017. Slides available at https://www.kleinerperkins.com/
perspectives/internet-trends-report-2017
Moore, R. L. (1994). Selling God: American religion in the marketplace of culture.
New York, NY: Oxford University Press.
Muslim Travel Girl. (n.d.). About me. https://muslimtravelgirl.com/about-me/
Nielsen. (2018). Time flies: U.S. adults now spend nearly half a day interacting with
media. https://www.nielsen.com/us/en/insights/news/2018/time-flies-us-adults-
now-spend-nearly-half-a-day-interacting-with-media.html
O’Loughlin, M. (2017, December 29). Following Francis: The pope’s social media min-
istry takes off. America Magazine. Retrieved from https://www.americamagazine.
org/faith/2017/12/29/following-francis-popes-social-media-ministry-takes
Public Citizen. (2017). Investigation shows that FTC’s reminder letters are ineffective
at disclosing paid posts on Instagram. Retrieved from https://www.citizen.org/
news/investigation-shows-that-ftcs-reminder-letters-are-ineffective-at-disclos-
ing-paid-posts-on-instagram/, https://commercialfreechildhood.org/investiga-
tion-shows-ftcs-reminder-letters-are-ineffective-disclosing-paid-posts-instagram
Twitchell, J. B. (2004). Branded nation: The marketing of megachurch, college inc., and
museumworld. New York, NY: Simon & Schuster.
United Methodist Communications. (n.d.). Rethink Church. http://www.umcom.org/
rethink-church
Valentino-DeVries, J., Singer, N., Keller, M. H., & Krolik, A. (2018, December 10). Your
apps know where you were last night, and they’re not keeping it secret. The New York
Times. https://www.nytimes.com/interactive/2018/12/10/business/location-da-
ta-privacy-apps.html
Wineburg, S., McGrew, S., Breakstone, J., & Ortega, T. (2016). Evaluating informa-
tion: The cornerstone of civic online reasoning. Stanford Digital Repository. Retrieved
from http://purl.stanford.edu/fv751yt5934
Wojdynski, B. W., & Evans, N. J. (2015). Going native: Effects of disclosure position and
language on the recognition and evaluation of online native advertising. Journal of
Advertising, 45(2), 157–168.
9.
Audiences, Affects,
Attachments: Theorizing Textual
Approaches to Digital Culture
Akane Kanai
Not long ago, a reviewer of the manuscript of my first book raised a long-
standing methods question for me. In that book, I explored young women’s
use of GIFs and captions on Tumblr to produce humorous moments for gen-
eral circulation. These were reaction-GIF posts, spanning experiences such as
trying but failing to flirt with cute guys, the attachment to continually eating
carbs, and feelings of contempt for girls wearing velour tracksuits. Following
the work of Margaret Wetherell (2012), I employed an affective-discursive
analysis to analyze how these posts invited imagined readers into a digital
intimate public, sharing feelings of embarrassment, struggle, and glee that
were positioned as relatable in certain gendered, classed, and racialized ways.
In the vein of feminist audience research, I explored how these young women
used, interpreted and refashioned media, and mediated discourses of feminin-
ity to produce these affective everyday moments. In the detail of the review,
what struck me were two related observations. The reviewer suggested that
although the book spoke to the priorities of feminist media studies, I had
followed a “well-trodden path” in my textual analysis, and it made an insuffi-
cient contribution to the field of social media research. I mention this not to
dispute the reviewer’s evaluation, which was both thoughtful and construc-
tive. Rather, I highlight its dialectical connection with anxieties, both mine
and those of other scholars, over what counts as “real digital media research.”
In arguing that attending to texts did not advance social media research, the
reviewer touched on some key tensions regarding how to adapt methods for
digital culture, and the place of textual approaches in understanding the sig-
nificance of media as a cultural phenomenon.
140 Akane Kanai
In this chapter, I address some of the important and distinctive uses a fem-
inist audience-text approach can offer in understanding affective practices of
sociality and identity in digital culture. I suggest that the text-reader metaphor
(Livingstone, 2004) is one useful way of understanding audiences, in terms of
analyzing the connections between subjectivity and the attachments to identi-
ties that discourses make available to us. I explore the case study of the grow-
ing body of research on digital intimacies and the influence of theorist Lauren
Berlant in terms of feminist and queer perspectives. Via Berlant, I suggest that
the premise of much contemporary research on digital identity and its entangle-
ments with new forms of intimacy owes its power to the notion of audiences’
affective engagement with texts and thereby discourse and collective meaning.
I further argue that a textual approach, among other approaches, is
needed in a moment where identities are increasingly textualized in digital
culture. Focusing on celebrity and its use and reception, I chart the con-
tinuing importance of celebrity as public text and the negotiation of nor-
mative gender. In this context, I discuss the ways in which texts are used by
participants in digital culture to both anchor and remake identity using the
case studies of Inger-Lise Bore’s (2017) analysis of the pinning of comedian
Amy Schumer on Pinterest, and my own research of the remixing of actress
Jennifer Lawrence’s star-image on Tumblr. Accessing these DIY texts pro-
vides a means of understanding the continuing compulsoriness of so-called
authenticity in contemporary femininity, as its abject Other, “artifice,” con-
tinues to be a specter haunting feminine self-representation in digital culture.
I suggest the selective incorporation of celebrity public texts into women’s
own (textual) narrations of selfhood shows the adoption, curation, and exci-
sion of particular femininities. Thus, texts can continue to tell us about what
forms of subjectivity are mandated and workable.
In making these points, I am not arguing that textual analysis is necessary
for every kind of inquiry. A text-reader approach does not prevent the use of
other methods such as interviews or ethnography as means of getting at how
audiences make sense of the media they use and consume. My argument is
that the text-reader metaphor (Livingstone, 2004) allows a particular way
of getting at questions of power in examining changing identities in digital
social spaces. I suggest that an approach taking the audience-text relation
seriously provides a means of accessing unequal and differentiated affective
negotiations of participants in digital culture.
Audiences, Users and Intimate Publics

There have been vigorous debates over what has been termed the seman-
tics of digital culture research (Das, 2010), given the attached presumptions
Affective Attachments, Audiences and Digital Texts 141
of activity, passivity, change, and stasis. Among other terms, these debates
have centered on the concepts of user, produser, and audience. Questions
over whether the user or audience are viable heuristics stem from longstand-
ing debates over who uses or is used (or both) in digital culture, and how
use is related to reception. Although the dominant term “user” is perhaps
unavoidably convenient in discussing participants of different platforms, José
van Dijck (2009) has questioned the cultural, economic (marketing), and
labor perspectives that uncritically celebrate the user, usually in terms of cre-
ative collaboration, producing content, or amateur skills. However, like Bird
(2011), who asked, “[a]‌re we all produsers now?,” van Dijck questioned the
empirical accuracy of the user as generalized description, noting that user
suggests action, more than the implicitly passive consumer of old media.
Audience researchers, indeed, have argued for active audiences for some time,
in that audiences were never exactly silent or passive (Das, 2010) and van
Dijck made a well-known observation that the active commenter only con-
stitutes about 10% of engagements with online content, with most preferring
to read or lurk.
The continuing desire to claim the agentic personhood of the user—priv-
ileging acting over reception or consumption—may lead to obscuring many
of the earlier contributions of older work on audiences. Livingstone (2004)
argued some time ago that:
We seem to treat ‘the Internet’ as a ‘black box’, despite having developed a

complex theory of codes, genre, mode of address, etc. for analysing television.
Tacit assumptions are made about Internet users—their interests, thoughts and
choices—as if we never found it necessary to study empirically the implied and
actual readers of television texts (p. 80).
For Couldry (2011), there is a renewed need for audience research that
is termed audience research as such, in order to understand what people are
doing around and with media, embedded in the practices of everyday life.
Indeed, texts, as I explore here, provide a particular route to understanding
audiences’ interpretations of the identity resources that are available to them.
Despite continuing changes in terms of the forces of convergence in the con-
temporary media environment, or what has been termed a media manifold,
Livingstone suggested that the emphasis on interactivity in digital culture puts
“interpretative activities at the very centre of media design and use” (2004,
p. 78). In such times, whereas the notion of the audience as we know it was
brought into existence by the dominance of mass media institutions, Couldry
(2011) was skeptical whether the digital era, characterized as it is by continu-
ing concentrations of power, has brought about significant enough change to
abandon existing lessons learned from an audience-based media culture.
142 Akane Kanai
Livingstone (2004) noted that because media are further interconnected

or, in other words, converged, it is content regardless of medium that matters
to its followers, and content-related concepts such as genre and discourse
still matter. Thus, the notion of audience is still important in studies of dig-
ital culture; it affords a means of understanding participation in collective
forms of meaning, or the discursive ties that enable belonging and exclu-
sion to occur. Livingstone’s observations resonate with analyses such as that
of Nancy Thumim (2012) who utilized the notion of genre to understand
digital self-representation. According to Thumim, self-representation takes
place within a set of expectations or tacit meanings between producer and
audience, whose roles, although often interchanged or collapsed in digital
contexts, still may be analytically distinct.
Livingstone argued that use and reception from an audience perspec-
tive have not always been so disconnected. To a certain extent, digital cul-
ture’s convergent movements work to combine both functions. Casting social
media participants or users as audience members resituates them as actors
who are always already imbricated not only within social customs but also
within media culture, as they participate in, consume, and remake media.
Thus, the bid to understand digital culture through the lens of the text-reader
metaphor encompasses both interpretive and everyday activities; texts consti-
tute a means of accessing the ways in which audiences are both “interpreters
of the media-as-text and users of the media-as-object” (Livingstone, 2004,
p. 83). Indeed, as I elaborate below, participants in social media are invited to
engage textually in such spaces, understanding the self and others as texts that
are open to be read. By using the terms “text” and “reader,” I refer to broadly
social interpretive functions that are vital to understanding the dynamics of
social digital spaces through an audience lens. Thus, the text-reader metaphor
is a means of conceptualizing users as already participating in, or hailed by,
existing forms of discourse as members of a loose audience.
In studies of social media culture, the notion of the audience has been
acknowledged as important but somewhat unevenly taken up. In Marwick
and boyd’s (2011) exploration of what they called the “imagined audience,”
they suggested that participants’ self-presentation is shaped by the mediated
realities of having a large and unknowable audience in the context of one-
to-many communication. Marwick and boyd highlighted audiences’ role in
re-shaping the expectations of the Tweeters in the study. Thus, the audience
features only as an imagined collectivity to whom the produser microcasts.
In a media-saturated context, I suggest it is useful to extend Livingstone’s
text-reader metaphor to loosely conceptualize the user or micro-caster also as
a member of a particular audience—a participant in discourse and collective
forms of meaning. In a culture that facilitates media-saturated practices by

consumers (Bird, 2003), users are always already audiences who make invest-
ments and divestments in broader forms of collective meaning and social
structure. This duality is one of the strengths of older contributions to audi-
ence research, in understanding how audiences make do with the discursive
and affective resources available to them.
An emphasis on collective participation in discourse, and practices pred-
icated on discourse, can be usefully mobilized through the text-reader met-
aphor. As a case study for this text-audience relation, I take the emerging
area of digital intimacies research that interrogates both the way in which
digital cultures produce new forms of intimacy and how such intimacies are
negotiated by individuals. Andreassen, Petersen, Harrison, and Raun (2018)
noted that there are two dominant threads in digital intimacies research.
One derives from the work of sociologist Ken Plummer (2003), focusing on
de-traditionalization and complexification in intimate relationships as a result
of late modernity. The other is inspired by Lauren Berlant’s articulation of
intimacy as a normative and regulatory relation (2008, 2011). In contrast to
Plummer, who focused on transformations in public/private divides from the
traditional to the late modern, Berlant questioned the divide of public and
private from a queer and feminist perspective, on the basis that private desires
already emanate from public discourse. As such, for Berlant, intimacy in the
sense of an intimate public does not, as Plummer suggested, arise through
private desires entering the public domain through the practices of extended
disclosure. The “autobiographical isn’t the personal” as Berlant (2008, p. vii)
neatly noted.
Paralleling the tensions and distinctions between the user and audience,
the distinctiveness of Berlant’s contribution to digital intimacies research is
premised on an understanding of audiences, texts, and the relation of col-
lective fantasy and desire. In The Female Complaint, Berlant (2008) wrote
of the “tender fantasies of the good life” (p. 1) promulgated by women’s
genres across literature and film. The affective structures and their magne-
tism for feminine audiences identified in the Female Complaint are drawn
from melodrama (Harriet Beecher Stowe’s Uncle Tom’s Cabin, the stage and
screen adaptations of Edna Ferber’s novel, Show Boat, and the film, Imitation
of Life), in which Berlant highlighted the promise of normalcy and belong-
ing. The texts that feature prominently in Berlant’s analysis play a central
and necessary role in the identification of everyday desires that circulate in
intimate publics. Berlant privileges the collective fantasies required in media
circulation; fantasies that are not individual but are the result of continual
circulation in mass media; fantasies distilling histories of power and difference
144 Akane Kanai
that allow women to feel general, average, or, as I have argued, relatable
(Kanai, 2019). The intimate public gives “permission to live small but to
feel large; to live large but to want what is normal too; to be critical without
detaching from disappointing and dangerous worlds and objects of desire”
(Kanai, 2019, p. 3).
The notion of the digital intimate public in this genealogy, then, draws
on a history of calls to collective identity via the mass media circulation of
texts. Warner similarly argued that a public is a space for the “reflexive circu-
lation of discourse” (2002, p. 62) in which participants may read themselves
into a shared social imaginary. The circulation of texts, then, ranging from
the Spectator (Warner, 2002) to Uncle Tom’s Cabin (Berlant, 2008) to reac-
tion-GIFs on Tumblr (Kanai, 2019), say much about a social imaginary and
one’s place in it. To Warner, this textual circulation is thus centrally impli-
cated in the production of social relations. In his case study of the English
serial magazine The Spectator, Warner argued that its circulation produces and
reveals “a particular culture, its embodied way of life, its reading practices,
its ethical conventions, its geography, its class and gender positions, and its
economic organization” (2002, p. 76).
An analysis of an intimate public, following Berlant, should therefore say
something about audiences and their social positioning in relation to particu-
lar discourses in their circulation often through texts. I suggest that Berlant’s
focus on texts allows the exploration of such investments in conventional-
ity, normalcy, and hopes of the good life. Such an approach highlights the
“structure of feeling” (Williams, 1961, p. 48) through which audiences are
addressed. As Ien Ang (1985) argued in relation to the soap opera Dallas,
fictional texts provide spaces for play and for trying out emotional vantage
points that are circumscribed in day-to-day contexts. Dallas did not operate
as a discrete object out there but as an everyday form of fantasy that women in
particular could use to adopt and try out positions “without having to experi-
ence their actual consequences” (p. 134) or their reality value. As such, texts
can provide a means of thinking through the tension between expectations
and practices and the affective attachments to particular forms of identity that
are not possible. They are a convenient point of departure from which to
think about genre—the sets of expectations through which individuals partic-
ipate in forms of culture (see also Thumim, 2012).
Moving from this genealogy to contemporary digital intimacies research,
Berlant’s work marks the publicness of the circulation of intimate affect
among a collectivity (Dobson, Carah, & Robards, 2018), ranging from jeong
in the case of the recovery of mobile phone footage captured by victims of a
capsized ship at sea in Korea (Hjorth, 2018), to the circulation of collective
fantasies of romance safely couched within so-called normal terms (Evans

& Riley, 2018). Berlant’s intimate public is used to describe the close-knit
masculine communities of gamers (Elliot, 2018) and generic practices of
oversharing on platforms such as Facebook (Kennedy, 2018). In particular,
Dobson et al. (2018) interpreted Berlant’s view of the intimate public as “cul-
tural scenes that promise and generate feelings of belonging and consolation”
(p. 5), citing Berlant’s description of the intimate public as a “circulation
of texts and things that express those people’s particular core interests and
desires” (2008, p. 5).
However, I suggest that although these operationalizations of Berlant
often draw on digital culture in which certain forms of disclosure occur,
Berlant argued that such disclosures do not necessarily produce intimacy.
Rather, intimacy is a centering of genre, discourses, and attachments to par-
ticular discursive positions that are available. I underline Berlant’s articula-
tion of the general as underpinning the sense of belonging in an intimate
public, rather than the avowal of idiosyncratic circumstances in and of them-
selves. As such, a public is intimate in Berlant’s imagining, not necessarily
because of acts of sharing or disclosure by its participants, but because of the
promise of belonging in a conventional way that inheres in generalizable dis-
courses of love, loss, and hope. In short, intimacy in Berlant’s view is linked
to already-existing discourses, and how audiences are positioned by them.
Berlant’s concept of “intimate public” foregrounds the audience and gives
a means of understanding how participants in everyday digital participatory
culture make use of the discourses through which they are already addressed.
Thus, texts are often central to the understanding of identity-making as part
of a shared digital intimate public.
I pause here to note, following feminist scholars of affect and emotion
such as Sara Ahmed (2004), Arlie Hochshild (1983/2003), Clare Hemmings
(2005, 2012), and Margaret Wetherell (2012), as well as the work of Nathan
Rambukkana (2015) on intimate digital counterpublics, that in using the
term “discourse” to address the question of intimacy and affect, I do not
see these as oppositional concepts. The notion of being addressed, invited,
or hailed into a particular discursive genre (I am also thinking here of Stuart
Hall’s (1996) theorization of identity) provides a way of addressing the
shared imaginary that is so vital to Berlant’s and Warner’s analyses of pub-
lics as affective spaces. Signification and affect are often closely bound up
together in the work of binding people to particular situations. Further, in
a pragmatic sense, digital cultures often privilege the visual, the textual, and
the affective simultaneously. As I have argued in relation to the reaction-GIF
cultures on Tumblr, such cultures collapse neat boundaries between affect,
146 Akane Kanai
discourse, text, and image. Thus, a continued emphasis on the mutuality and
the dynamic interaction between text and reader, or alternatively, discourse
and audience, is crucial to understanding how affective boundaries in digital
publics are made and remade.
The lens of the audience-text relation provides a particular means of
understanding how individuals participate in affect worlds that derive from
collective forms of meaning. Such a focus need not mean a lack of attention
to the specificities of the digital cultures and platforms through which such
participation takes place. Indeed, I suggest that the text-reader metaphor may
be used to position the platform as a political player in the mediation of
content and the reconfiguration of audiences and other social formations.
Platforms, as Gillespie (2015) noted, often present themselves as neutral, flat,
and empowering spaces, an even ground through which users may unprob-
lematically amplify their voice—a view that correlates more with readings of
the Habermasian public than with Berlant’s intimate public (Ferreday, 2018).
Attention to the discursive conditions that constrain, incite, and shape the
sociality of platform cultures is also a means of attending to platform specific-
ity. I note there is much excellent work in this vein—for example, in feminist
work such as Amy Dobson’s (2015) emphasis on the over-surveilled nature
of youthful femininity across platforms; Jessalynn Keller’s, Kaity Mendes’,
and Jessica Ringrose’s (2018) research that explores how feminist girls and
women experience Twitter; and Katrin Tiidenberg’s (2015) research on the
use of sexy selfies on Tumblr.
My aim here is not to reinvent the wheel, but to highlight the potential
of drawing links to older work on audience engagement with media texts as
living practices. A textual analysis is not something that must be used but
rather has a specific place in understanding digital culture. In a convergent
context in which references to television, film, and celebrity culture are mixed
in with native forms of digital content, a discursive analysis that is attentive to
texts is at times necessary for explorations of participants’ collective fantasies,
investments, and attachments, particularly in terms of gender.
Textualization: Identities, Celebrity and Everyday Practices

So far, I have argued that texts are a distinct but crucial means of approaching
how audiences may attach to or reject particular discursive positions. In this
section, I suggest that an awareness of textual approaches becomes even more
important as identities in digital culture are increasingly textualized. As many
feminist scholars of identity have noted, girls and young women face inten-
sifying expectations to be visible and produce the self in a highly disciplined
fashion in online contexts (Bailey & Steeves, 2013; Brandes & Levin, 2013;
Dean & Laidler, 2013; Harris & Dobson, 2015; Rossie, 2015) in conjunc-
tion with neoliberal imperatives to demonstrate value across professional,
social, and personal contexts. Much of this feminist research highlights that
young women in particular increasingly understand in their work of profile
curation that they are viewed and read as texts. In contemporary post-Fordist
culture whereby individuals must increasingly perform and display their value
(Skeggs & Loveday, 2012), celebrity and its relational logics have moved
beyond the wealthy and famous to become part of common sense gram-
mars of digital identity practice (Marwick, 2013). As digital culture promotes
logics of visibility (Bucher, 2012), audiences themselves are both incited to
perform celebrity-like forms of curation and required to develop literacies to
evaluate, discern, and perform everyday forms of textual analysis in relation
to the mediated personalities they consume and use. As such, we may observe
that the production of the self as text, inviting particular forms of reading,
increasingly constitutes an everyday form of social practice.
I initially came to thinking about textualization through my research
on Tumblr. I found that young women creatively repurpose GIFs to make
notionally self-representative feelings intelligible, and legible, to unknown
publics. Importantly, such feelings did not have a necessary link to the offline
or reality. As one of my participants stated, it was of little relevance whether
she was in bed watching Game of Thrones while she released a post about
being drunk at the bar; what mattered was that the post was funny and
appealed to her audience. These young women on Tumblr demonstrated the
capacity to discern and perform the normative averageness of White middle
class youthful femininities: not perfect but, as Berlant would say, in proximity
to the good life. This was clear in how the chosen themes were narrated in the
posts: not matching one’s own (high) expectations; continually embarrassing
oneself; being frustrated by needy girlfriends; desiring “hot guys” but never
quite mastering flirtation. The reaction-GIF texts distilled self-representative
experience that, in their humorous, relatable, and sociable affective tenor,
were amenable to further circulation on Tumblr.
In what follows, I outline how the logic of textualization operates using
women’s relation with female celebrity as a paradigmatic example of these
movements in digital culture. Celebrity has long been understood as a public
text (Dyer, 2004). The examples I discuss of the reception/use of Jennifer
Lawrence and Amy Schumer show how women draw on female celebrity as
a textual resource, as both audiences of celebrity and participants in social
media cultures that operate according to celebritized logics. In the blurring
of boundaries between female celebrities as public texts and women’s digital
148 Akane Kanai
labor in representing themselves, we may observe the remaking of the signif-

icance of celebrity women in distilling particular desired and legible elements
of femininity, and erasing others. These digital textual remakings by audiences
further reveal connections between celebrity, identity, and culturally idealized
and suppressed forms of subjectivity. Here I highlight the further malleability
and proximity of female celebrity texts used as markers of relatability, aspira-
tion, differentiation, or disgust that are spliced in with the self-representative
texts of ordinary women.
Celebrity Texts and Self-Textualization

Even before the dominance of digital culture, celebrity selfhood has exem-
plified the production of self as public text and been analyzed as a means
of understanding power, identity, and the relations between the individual
and capitalist production (Marshall, 1997). Feminist scholars have produced
much important work on the function of female celebrity for desirable and
abject femininities and celebrity identities. As Jackie Stacey (1994) found in
her groundbreaking research on female audiences and their relation to cinema
in 1940s Britain, celebrity has been a key way in which notions of desirable
femininity are established and gain affective force and traction in public audi-
ences. Observing the practices of consuming glamorous Hollywood feminin-
ity in which women audiences participated at the time, Stacey’s work showed
the entanglement of everyday practices, such as mimicking the couture of the
stars, with the affective reception of these filmic texts.
The continuing and intensified salience of celebrity indicates that an
everyday form of textual analysis of celebrity can help establish and make sense
of the parameters of individuality in contemporary life. Circulated in media
channels spanning the cinema, TV appearances, and gossip cultures, celebrity
texts have long been open for re-making by industries and fans. In digital
culture, as the self-representations of both celebrities and ordinary people
move beyond the original contexts of their articulation, I suggest an everyday
form of textual analysis has become mundane. In contemporary media cul-
ture, as star images travel and morph across media platforms including digital
platforms, consumers of celebrity media—fans, antifans (Gray, 2003), and
observers in general—are visibly producing traces of celebrity reception and
use via comments, remix, and other user-generated content. This content not
only demonstrates how certain aspects of the celebrity public text are seen but
also how certain identities are made significant in popular memory.
In my Tumblr research (Kanai, 2015), I observed these movements in
the remixing of Jennifer Lawrence through the creative and selective use of
GIFs. Unlike much research on participatory cultures that focuses on fandom

of a particular celebrity or media artifact as a way of doing audience research,
I focused on identity and its negotiation through the reaction-GIF contexts
and somewhat incidentally found consistent themes in the use and re-mix-
ing of Jennifer Lawrence’s star text. This consistency was notable given the
miscellaneity of the GIFs used by these young women. The GIFs featured a
variety of scenarios featuring dogs, cats, cars, famous and less famous humans
from reality television, film, and YouTube videos in self-representative and
fictional roles. These actors were essentially ventriloquized in that they were
used to express the feeling or action articulated by the Tumblr user.
Notably, Lawrence—unlike other film and television actors featured in
the GIFs—was usually featured as herself rather than one of her fictional char-
acters; the GIFs often used interview or red carpet and candid moments.
As a slim, able-bodied, wealthy, young White woman who won an Oscar
at a young age, and who regularly features in haute couture branding cam-
paigns, Lawrence’s celebrity script could legitimately be considered relatively
unachievable. Despite this, Lawrence’s understood persona as a seemingly
“normal” young woman was a clear theme in which she was reconstructed and
used to signify the Tumblr user’s own relatability. Lawrence was understood
and mobilized both in terms of authenticity and her capacity to play with the
rules of femininity. Most commonly, this was articulated via the classed, racial-
ized ability to not quite meet social norms for women, while being forgiven
and even celebrated for these breaches. Self-deprecating quotes such as “I’m
not cool enough to be at this party” and “I have the street-smarts and survival
skills of like, a poodle” featured heavily in the GIFs.
Thus, Jennifer Lawrence’s text was condensed and distilled to not only
manifest the presentation of normative femininity in a seemingly authentic way
and natural way, but also erase the labor associated with doing so. Indeed,
it is important to observe that normative demands of the naturalness and
authenticity of femininity have intensified in a context where the configura-
tion of neoliberalism and existing gender inequalities produces the increasing
compulsoriness for girls and young women to invest labor in their identities,
and continually demonstrate their value (Banet-Weiser, 2015). Young women
in particular are required to be visible in highly disciplined ways, through the
managing of narrow parameters of legitimacy. Young women must produce
visual and textual self-representations that evince authenticity without over-
sharing, attracting likes without overtly seeking them out (Rossie, 2015).
In short, girls and women are required to think carefully about their online
texts and how they may be read, re-read, and perhaps willfully read against
the grain.
150 Akane Kanai
Inger-Lise Bore’s (2017) insightful analysis of the pinning of Amy

Schumer on Pinterest similarly demonstrates how social media participants
engage in collective fantasies of normative femininity via the reconstruction of
female celebrity as public text. Schumer is popularly understood as an “unruly
woman” (Rowe, 1995), particularly for her sexually transgressive comedy.
Bore argued that as a White American female comedian, Schumer both
appropriates and dismantles the male gaze in her humor centered on cultural
constructions of sex and the female body; she also is well known for her obliv-
iousness to, and at times engagement in, casual racism (Patton & Leonard,
2015). Bore demonstrated that, in the otherwise unstable and contradictory
public text attached to Amy Schumer, relatively consistent themes emerged
in the top 100 pins of Schumer on Pinterest: the dominant reconstructed
narratives depicted Schumer as an aspirational, even glamorous, female celeb-
rity, rather than as a comedian. Further, Schumer’s problematic history with
race was invisible, matching what might be understood as the feeling rules
of the platform. Pinterest, as Bore noted, is a platform noted for its cultures
of domesticity, aspirational lifestyle and luxury, and commitment to self-im-
provement; it is thus likely to encourage participation by those who already
have access to privilege.
Schumer was used as a set of symbolic resources in ways that demonstrated
her resonance as an inspirational figure, who offered ways to valorize the self
and gain confidence and self-acceptance. Quotes of Schumer’s such as “I say
if I’m beautiful, I say if I am strong, You will not determine my story—I will”
were overlaid on images of nature, or serious (as opposed to comic) images of
Schumer. Authenticity emerged as a dominant resignification of Schumer—
although, as Bore observed, much of her celebrity text is marked by her ironic
performance of White femininity. Similarly, Schumer’s body was the focal point
in these pins, but often in paratexts privileging celebrity glamor rather than her
comedic observations that often focused on the unruliness of her own body
and sexual transgression. As such, Schumer’s celebrity image was stabilized,
largely positioned within a project of aspirational self-development.
Thus, the use of Schumer fits with broader observations of the political
imperatives that require the self to become the center of investment in neo-
liberal culture. Branding, as Sarah Banet-Weiser (2012) noted, has become
the logic through which citizens affirm their worth. Branding is different
from commodification; for Banet-Weiser, it instantiates a different set of
social relations, “a dynamic between production and consumption, between
the individual and the culture at large,” that requires the work of adapting
and innovating the self while ideally producing a “unique, authentic self”
(p. 73). Thus, although branding in digital culture has become a normative
means of demonstrating value, it raises questions about the politics of the

relation between the brand and audience. Besides the simplification of iden-
tities implicated in branding, what kinds of vulnerabilities and inequalities are
glossed over in this relation of production and consumption?
If we accept that bodies are continually differentiated in terms of their
value, it follows that scholars need to attend to issues of power in the everyday
mechanisms through which identities are distilled into textual resources for
self-expression. How does re-reading and the re- and de-contextualization
of selves work in a medium that enables persistence or permanence, search-
ability and replicability, and in a context where audiences are often presumed
but unknown and invisible (boyd, 2007)? There are broader questions here
in relation to how minoritized populations are used, re-used, and re-made
through social media participation. For example, Lisa Nakamura (2014) ana-
lyzed the ordinary violence enacted on racialized people when they are con-
verted into online images in scam-baiting meme sites. Indeed, even seemingly
non-violent and inoffensive forms of de-contextualization can have violent
effects dependent on race (Kanai, 2016). In this vein, Aria Halliday (2018)
argued that the textual circulation of celebrity on social media can contribute
to the ongoing forms of racialized violence suffered by Black women. Taking
rapper Nicki Minaj as a case study, Halliday documented the Instagram
remaking through which Minaj’s body was taken apart to highlight her but-
tocks but remove her head, replaced with that of other pop culture figures: a
textual beheading that fits into the broader mosaic of violence against Black
feminine bodies, read as amenable to White desires.
Textualization, then, is much more than simply a practice of representa-
tion understood as mirroring or reflecting reality; rather, these textual prac-
tices show us how digital culture enables it as an everyday form of social
practice—extending, affirming, simplifying, and enacting violence on iden-
tities. In digital culture, texts indicate the ways in which valuable subjectivi-
ties are curated, distilled, and remixed from the discursive resources available
to them. In everyday practices of textualization in digital culture, selves are
transformed into narratable objects. Thus, while much of the textual remix of
social media can be seen as a form of play and sociability, questions also arise
in relation to the violence that also may perpetuate and re-stabilize already-ex-
isting inequalities in the plastic reconfiguration of bodies as texts.
Audiences, Texts, and Power

I have sought to clarify some of the uses of textual methods based on femi-
nist priorities and perspectives, and suggested that texts afford a distinctive
152 Akane Kanai
way of accessing audiences’ affective investments in particular forms of iden-

tity and collective meaning. Following Livingstone (2004), digital texts may
be approached as the traces of social practices that demonstrate both the
reception of particular understandings (media-as-text) and uses (media-as-
object). The flourishing of digital intimacies research is heavily indebted to
Lauren Berlant (2008, 2011), but I have argued that this is largely because
of Berlant’s privileging of the audience-text relation and the affective magne-
tism of particular collective meanings that are used as resources for everyday
hopes, attachments, and identity-making.
Texts are not the only means of approaching audiences. Yet, as I have
argued here, practices of textualization are increasingly mundane, common,
and instinctive ways of doing identity in digital culture. The blurring of
boundaries between the circulation of celebrity public texts and everyday
identity practice is one example of how texts continue to constitute paradig-
matic means of articulating legible femininities, whether by traditional audi-
ences or highly mediated personalities. Yet, texts also enable certain ways of
responding to identities; indeed, the textualization of the self creates risks
of re- and de-contextualization in the re-readings made possible through
digital circulation. As a text in contemporary digital attention economies,
the self may be simplified, hollowed out, and taken apart in gendered,
classed, and racialized ways. Further questions are posed by these move-
ments, in terms of how such expanded forms of textualization work given
the existence of continuing inequalities—who is reified, pinned, mobilized,
and with what consequences? This chapter has attempted to sketch out the
basis for continuing inquiries into digital culture that foreground affective
attachments, practices, and collective meaning as a means of approaching
these questions.
References
Ahmed, S. (2004). The cultural politics of emotion. Edinburgh, UK: Edinburgh
University Press.
Andreassen, R., Petersen, M., Harrison, K., & Raun, T. (Eds.). (2018). Mediated intima-
cies. London, UK: Routledge.
Ang, I. (1985). Watching Dallas: Soap opera and the melodramatic imagination (D.
Couling, Trans.). London, UK: Methuen & Co.
Bailey, J., & Steeves, V. (2013). Will the real digital girl please stand up? In J. M. Wise &
H. Koskela (Eds.), New visualities, new technologies: The new ecstasy of communication
(pp. 41–66). Farnham, UK: Ashgate Publishing Ltd.
Banet-Weiser, S. (2012). Authentic TM: The politics of ambivalence in a brand culture.
New York: New York University Press.
Banet-Weiser, S. (2015). ‘Confidence you can carry!’: Girls in crisis and the market for
girls’ empowerment organizations. Continuum, 29(2), 182–193.
Berlant, L. (2008). The female complaint: The unfinished business of sentimentality in
American culture. Durham, NC: Duke University Press.
Berlant, L. (2011). Cruel optimism. Durham, NC: Duke University Pres.
Bird, S. E. (2003). The audience in everyday life: Living in a media world. London,
UK: Routledge.
Bird, S. E. (2011). Are we all produsers now? Cultural Studies, 25(4–5), 502–516.
Bore, I.-L. K. (2017). Screen comedy and online audiences. London, UK: Routledge.
boyd, d. (2007). Social network sites: Public, private or what? Knowledge Tree, 13. Available
at https://www.danah.org/papers/KnowledgeTree.pdf
Brandes, S. B., & Levin, D. (2013). Like my status. Feminist Media Studies, 14(5),
743–758.
Bucher, T. (2012). Want to be on the top? Algorithmic power and the threat of invisibility
on Facebook. New Media & Society, 14(7), 1164–1180.
Couldry, N. (2011). The necessary future of the audience … and how to research it. In
V. Nightingale (Ed.), The handbook of media audiences (pp. 213–229). Chichester,
UK: Wiley-Blackwell.
Das, R. (2010). Meaning at the interface: New genres, new modes of interpretative
engagement? The Communication Review, 13(2), 140–159.
Dean, M., & Laidler, K. (2013). A new girl in town: Exploring girlhood identities through
Facebook. First Monday, 18(2). Retrieved from http://journals.uic.edu/ojs/index.
php/fm/article/view/4230
Dobson, A. S. (2015). Postfeminist digital cultures: Femininity, social media, and self-rep-
resentation. New York, NY: Palgrave MacMillan.
Dobson, A. S., Carah, N., & Robards, B. (Eds.). (2018). Digital intimate publics and
social media. London, UK: Palgrave Macmillan.
Dyer, R. (2004). Heavenly bodies: Film stars and society. London, UK: Routledge.
Elliot, A. (2018). Speaking to the Other: Digital intimate publics and Gamergate. In A.
S. Dobson, N. Carah, & B. Robards (Eds.), Digital intimate publics and social media
(pp. 231–246). Basingstoke, UK: Palgrave Macmillan.
Evans, A., & Riley, S. (2018). ‘This dapper hotty is working that tweed look’: Extending
workplace affects on TubeCrush. In A. S. Dobson, N. Carah, & B. Robards (Eds.),
Digital intimate publics and social media (pp. 129–144). Basingstoke: Palgrave
Macmillan.
Ferreday, D. (2018). Something substantive to reach out and touch: The intimate politics
of digital anti-rape activism. In R. Andreassen, M. Petersen, K. Harrison, & T. Raun
(Eds.), Mediated Intimacies (pp. 19–32). London, UK: Routledge.
Gillespie, T. (2015). Platforms intervene. Social Media + Society, 1(1).
doi: 10.1177/2056305115580479
Gray, J. (2003). New audiences, new textualities: Anti-fans and non-fans. International
Journal of Cultural Studies, 6(1), 64–81.
154 Akane Kanai
Hall, S. (1996). Who needs identity? In S. Hall & P. du Gay (Eds.), Questions of cultural
identity (pp. 1–17). London, UK: Sage Publications.
Halliday, A. S. (2018). Miley, what’s good? Nicki Minaj’s Anaconda, Instagram reproduc-
tions, and viral memetic violence. Girlhood Studies, 11(3), 67–83.
Harris, A., & Dobson, A. S. (2015). Theorizing agency in post-girlpower times.
Continuum, 29(2), 145–156.
Hemmings, C. (2005). Invoking affect: Cultural theory and the ontological turn. Cultural
Studies, 19(5), 548–567.
Hemmings, C. (2012). Affective solidarity: Feminist reflexivity and political transforma-
tion. Feminist Theory, 13(2), 147–161.
Hjorth, L. (2018). Snapshots of afterlife: The cultural intimacies of posthumous camera
phone practices. In A. S. Dobson, N. Carah, & B. Robards (Eds.), Digital intimate
publics and social media (pp. 59–74). Basingstoke, UK: Palgrave Macmillan.
Hochschild, A. (1983/2003). The managed heart: The commercialisation of human feel-
ing. Berkeley, CA: University of California Press.
Kanai, A. (2015). Jennifer Lawrence, remixed: Approaching celebrity through DIY digital
culture. Celebrity Studies, 6(3), 322–340.
Kanai, A. (2016). Sociality and classification: Reading gender, race and class in a humorous
meme.’ Social Media + Society, 2(4). doi: 10.1177/2056305116672884
Kanai, A. (2019). Gender and relatability in digital culture: Managing affect, intimacy
and value. Basingstoke, UK: Palgrave Macmillan.
Keller, J., Mendes, K., & Ringrose, J. (2018). Speaking ‘unspeakable things:’ Documenting
digital feminist responses to rape culture. Journal of Gender Studies,27(1), 22–36.
Kennedy, J. (2018). Oversharing is the norm. In A. S. Dobson, N. Carah, & B. Robards
(Eds.), Digital intimate publics and social media (pp. 265–280). Basingstoke,
UK: Palgrave Macmillan.
Livingstone, S. (2004). The challenge of changing audiences: Or, what is the audience
researcher to do in the age of the Internet? European Journal of Communication,
19(1), 75–86.
Marshall, P. D. (1997). Celebrity and power: Fame in contemporary culture.
Minneapolis: University of Minnesota Press.
Marwick, A. E. (2013). Status update: Celebrity, publicity and branding in the social media
Marwick, A. E., & boyd, d. (2011). I tweet honestly, I tweet passionately: Twitter users,
context collapse, and the imagined audience. New Media & Society, 13(1), 114–133.
Nakamura, L. (2014). ‘I WILL DO EVERYthing That Am Asked’: Scambaiting, digital
show-space, and the racial violence of social media. Journal of Visual Culture, 13(3),
257–274.
Patton, S., & Leonard, D. (2015, July 6). Don’t believe her defenders. Amy Schumer’s
jokes are racist. The Washington Post. Retrieved from https://www.washingtonpost.
com/posteverything/wp/2015/07/06/dont-believe-her-defenders-amy-schum-
ers-jokes-are-racist/?noredirect=on&utm_term=.2a3df8a782ac
Plummer, K. (2003). Intimate citizenship: Private decisions and public dialogues. Seattle,

WA: University of Washington Press.
Rambukkana, N. (2015). From #RaceFail to #Ferguson: The digital intimacies of
race-activist hashtag publics. The Fibreculture Journal, (26). http://twentysix.
fibreculturejournal.org/fcj-194-from-racefail-to-ferguson-the-digital-intimacies-of-
race-activist-hashtag-publics/
Rossie, A. (2015). Moving beyond ‘Am I pretty or ugly?’: Disciplining girls through
YouTube feedback. Continuum, 29(2), 230–240.
Rowe, K. (1995). The unruly woman: Gender and the genres of laughter. Austin: University
of Texas Press.
Skeggs, B., & Loveday, V. (2012). Struggles for value: value practices, injustice, judgment,
affect and the idea of class. The British Journal of Sociology, 63(3), 472–490.
Stacey, J. (1994). Star gazing: Hollywood cinema and female spectatorship. London,
UK: Routledge.
Thumim, N. (2012). Self-representation and digital culture. Basingstoke, UK: Palgrave
Macmillan.
Tiidenberg, K., & Gomez Cruz, E. (2015). Selfies, image and the re-making of the body.
Body & Society, 21(4), 77–102.
van Dijck, J. (2009). Users like you? Theorizing agency in user-generated content. Media,
Culture & Society, 31(1), 41–58.
Warner, M. (2002). Publics and counterpublics. Public Culture, 14(1), 49–90.
Wetherell, M. (2012). Affect and emotion: A new social science understanding. London,
UK: Sage Publications.
Williams, R. (1961). The long revolution. London, UK: Chatto & Windus.
10.
Discourse-Analytical Studies on
Social Media Platforms: A Data-
Driven Mixed-Methods Approach
Ehsan Dehghan, Axel Bruns, Peta Mitchell and
Brenda Moon
As social media platforms have become ubiquitous in everyday life, calls for
mixed-methods approaches to studying social media platforms have also
increased. Especially following the so-called computational turn (Berry,
2011) in media and communication research, we have witnessed the devel-
opment of novel quantitative approaches, tools, techniques, and methods to
study social media platforms. Qualitative studies, in contrast, often continue
to draw from traditional methods that are sometimes not well equipped to
deal with the particularities of studying these spaces, including their affor-
dances, dataset sizes, systematic sampling, and downsizing of data, or simply
put, finding the right datasets.
This becomes particularly important for studies employing a discourse-an-
alytical perspective. Traditionally, selecting texts for discourse analysis has
been straightforward, with researchers using contextual knowledge to select
newspapers, articles, television programs, political speeches, and similar texts
for qualitative analysis. However, when it comes to social media, especially
if the researcher is interested in the broader discursive environment or the
everyday communicative practices of social media users, data selection and
downsizing becomes more complex.
We propose a mixed-methods approach to tackle some of the issues facing
discourse analysts investigating social media platforms, especially Twitter. In
doing so, we draw on social media analytics, network analysis, corpus lin-
guistics, and discourse studies. The proposed methodological pipeline equips
researchers with a series of metrics to identify social media research objects
158 Ehsan Dehghan, et al.
(posts, tweets, number of followers, visibility, and so forth) with salient dis-
cursive power; observe the underlying network structures and investigate
how the specific affordances of a platform influence the flow of information
and shaping of the discourse; and select an objective, downsized, and repre-
sentative sample of the data for further in-depth qualitative analysis.
We draw on examples from two different case studies to show how this
methodology can provide a richer understanding of the context, discursive
environment, and dynamics of communication on the platform. The case
studies investigate recent and prominent issues in the Australian socio-political
sphere: the controversial automated welfare-debt recovery scheme launched
by the Australian Government in 2016, which spurred the nationally promi-
nent #robodebt Twitter campaign; and the almost year-long discussions over
proposed changes to Australia’s Racial Discrimination Act in 2016–2017.
New Challenges for Discourse Analytics

Although the terms discourse and discourse studies encompass a wide array
of theoretical and methodological approaches, a common characteristic of
most, if not all, discourse-analytical studies is their emphasis on in-depth,
rich, and context-aware analysis. This often requires a focus on relatively small
datasets that are systematically selected and sampled to be representative of
the discourses and topics of the study, yet manageable under the time and
resource constraints of the researchers. In this regard, one key challenge for
discourse-analytical studies on social media platforms arises when making crit-
ical and systematic decisions on where to look. This is particularly important
for studies interested in the mundane, everyday discourses of social media
users. Unlike studies focusing on the discourse of politicians, news organiza-
tions, activists, and other distinct actors, in which the data collection often
follows a straightforward process starting from the identified accounts of such
individuals or organizations, the study of mundane social media discourses is
often faced with large volumes of data, scattered across the platform under
investigation. This necessitates the use of approaches that can account for
such challenges in data collection, selection, and sampling.
Further complicating discourse-analytical studies of social media activi-
ties are the platform-specific characteristics of social media discourses. In any
discursive environment, social actions are constrained by the material and
the discursive limits of the spaces in which they occur. In the case of a social
media platform, these constraints involve the logics of the platform itself
(van Dijck & Poell, 2013), including its technological design, affordances,
algorithmic curation and moderation practices, user demographics, perceived
Discourse-Analytical Studies on Social Media 159
and expected communicative culture, and other factors. These technical and
technological aspects of a platform constrain and constitute—and are con-
strained and constituted by—the discursive practices taking place on the plat-
form. At the same time, such technical aspects of a platform are themselves
rooted in the discourses of the designers, owners, content moderators, and, in
Foucauldian terms, the orders of the discourses in which the platforms were
created and within which they operate (Foucault, 1971). Social media analy-
sis, therefore, has much to gain from a discourse-analytical approach, but, as
noted above, long-established traditional approaches to discourse studies are
not well equipped with the methodological and analytical toolkits required to
account for such complexities.
Approaching these questions from a different disciplinary tradition, yet
facing many of the same challenges, is research generally rooted in media and
communication studies, and enhanced more recently with advanced compu-
tational methods. With the rise of the so-called participatory web, and with
it the huge leap in the volume of available data, new approaches in these
disciplines have attempted to accommodate the vast increase in what is called
“big data” or, more specifically, “big social data” (Halavais, 2015). But this
sudden enthusiasm for big data has at times led to an exaggerated focus on
the data themselves, without taking into account the origins and contexts
of the data sources (boyd & Crawford, 2012). The large volumes of data
collected from social media platforms have frequently been treated as ends in
themselves, leading to a substantial number of studies focusing on relatively
shallow analyses of patterns of interaction, collective sentiments, clustering
words and language, and the like. Interest in big social data has produced
great advances in managing and processing the sheer volume of data in a
limited time. However, it has also generated many studies that focus only
on exceptional events and phenomena (identified on Twitter, for instance,
by hashtags and other explicit discursive markers). Through their hashtag-
and keyword-centric data gathering strategies, such studies usually treat their
datasets as representative of isolated communicative spaces that are separate
from their everyday discursive environments—the surrounding social media
platforms and the media environments beyond the platforms themselves. In
extreme cases, this has led to media-centric and technologically determinist
moral panics blaming social media platforms for everything that is wrong with
society. One can see traces of such perspectives in the line of arguments—such
as those refuted by Bruns (2019)—about how the use of social media plat-
forms leads to polarization, filter bubbles, and echo chambers.
This argument, of course, should not be taken as completely dismissing
the impact of platforms on societal developments. As we have noted, the
materialities and embedded biases in the design, affordances, and governance

of platforms clearly influence discursive practices. But although the role of
these platforms in the everyday lives of users and non-users is undeniable, a
persistent focus on atypical, extreme examples of social media discourse can-
not provide a comprehensive picture of discursive practices on social media
platforms. Disregarding the socio-political contexts of discourses and solely
focusing on broad patterns of communication can lead to misguided conclu-
sions about discourses of and on social media.
As boyd and Crawford (2012) argued, regardless of the size of the data
available from a social media platform, it is always important to remember
that big data are created by humans, in human and temporal contexts, in the
context of platforms and algorithms created by humans, and are eventually
interpreted by humans. This intertwined nature of socio-political and plat-
form contexts means that studies of social media discourses must account
both for what happens on the platform and for the platform-external context
that shapes the discourses themselves. KhosraviNik and Esposito (2018) refer
to these as the horizontal and vertical contexts of social media discourses. On
the horizontal scale, we face the technological and discursive aspects of the
discourses that have shaped the platform, its affordances, and its operations.
On the vertical, there are issues of power, hegemony, socio-political contexts,
and other factors. At the intersection of the horizontal and vertical scales,
one can witness what Carpentier (2017) theorized as the “discursive–material
knot,” where no hierarchies are present between the discursive and the mate-
rial. Rather, the discursive (horizontal and vertical) and the material (techno-
logical design, affordances, and so forth) are co-present and co-constitutive
on a social media platform.
In a study of communicative and discursive practices on social media plat-
forms, therefore, there is always a need to account for the contextual pecu-
liarities of the datasets under investigation. Although big-data approaches
can provide valuable insights about the horizontal discursive environment,
communities of users, patterns of social actions on platforms, sentiments, and
the like, these insights need to be interpreted in light of their vertical contex-
t(s), the discourses that shape them, and discourses that are shaped by them.
Furthermore, any quantification of the material should be seen alongside the
discursive. Any large-scale, big-data analysis of social practices on platforms
has to incorporate a level of context-aware, cultural, and discursive analysis.
As Manovich (2012) put it, “a human is still needed to make sense of these
patterns” (p. 469). This calls for methodologies that can address large vol-
umes of data, incorporate platform-specific (horizontal and material) aspects,
and analyze discursive practices simultaneously.
Toward a Cyclical Mixed-Methods Social Media Discourse

Analysis
In this light, we propose a methodological pipeline that draws from both
quantitative, big-data approaches and qualitative, in-depth methods and
show how each of these can inform and complement the other. This meth-
odology draws from theories of discourse to exemplify how patterns of prac-
tice on social media platforms can be interpreted from a discourse-theoretical
lens. We will also show how big-data approaches can help a discourse analyst
identify where to look when faced with the sheer volume of datasets readily
available from social media platforms. The iterative, circular flow of move-
ment from big-data to small-data approaches, in which each step both feeds
into and is informed by the others, forms the backbone of this methodologi-
cal approach. Our approach cyclically moves between temporal and aggregate
metrics, social network analysis, and textual investigation of the datasets.
Data Capture
The first challenge for any researcher working with data from social media
platforms is selecting what data to gather. This is inherently shaped by the
available data access points—usually, the platform’s Application Programming
Interface (API) or alternative approaches, such as Web scraping—and by the
tools that connect to such access points. Although a wide variety of such tools
is available for common social media platforms, those tools often implement
similar data gathering approaches, which are in turn directly influenced by
the data structures that APIs and data access cost structures privilege. For
Twitter, the most common data-gathering techniques capture tweets contain-
ing preselected keywords or hashtags. Approaches that select tweets based on
user location, time zone, or other demographic factors, or whole-of-platform
approaches that do not preselect themes and populations a priori, are consid-
erably less common (Burgess & Bruns, 2015). It is imperative that research-
ers remain aware of how these methodological and political-economic factors
shape the data they work with and consider the impact of these biases on their
analyses.
Generic Social Media Analytics

Once the data are gathered, it is crucial to develop a bird’s-eye view of what
the dataset comprises. This requires the researcher to address a series of ques-
tions that emerge both from the logics of the platform and the discourses
shaping interactions on it. Especially central to studies collecting data for a
period of time (from a few days to a few years) is the question of the temporal
dynamics of interactions on the platform. Rises and falls in the number of
posts, tweets, comments, likes, and so forth provide clues about the news,
events, themes, and topics that have affected the patterns of participation.
These peaks and troughs in social media activities can point researchers to the
periods of time that are most interesting or relevant, whether researchers are
interested in the discourses that are represented, amplified, or foregrounded,
or those that are absent or backgrounded. Different patterns of activity over
time can reveal valuable information about the discourses and events that lead
to a higher level of reaction, or conversely about those that are ignored by
social media users.
The logics of platforms in terms of visibility are similarly important to
understanding activity levels. In general, the algorithmic design of platforms
often rewards forms of practice that increase user engagement. These include
having more followers, friends, or subscribers; posting, commenting, or
replying; frequently using reaction, retweet, like, upvote, and downvote but-
tons; and generally making more use of the platform’s affordances. Thus the
platform’s technological design and algorithmic moderation could affect the
discursive practices of its users.
Therefore, at the early stages of a study’s methodological conceptualiza-
tion, researchers need to consider these platform logics, by identifying the
objects that are more visible in one sense or another and reflecting on how
such visibility should be interpreted. Variations in the visibility of different
types of objects also reveal information about the discursive environment of
the study, in terms of the horizontal context of the platform and the vertical
context of the discursive environment. The affordances of platforms often
enact different discursive functions. On Facebook, for instance, one could
infer quite different conclusions about posts receiving high numbers of likes
and loves or angry and sad reactions. The same is true for the number of
retweets and @mentions received by different accounts on Twitter, the most
(or least) upvoted or downvoted posts on Reddit, or liked and disliked videos
on YouTube. Although these different affordances might be treated more or
less equally in the algorithmic moderation of content on the platforms, in that
they eventually create engagement, they reveal significantly different informa-
tion about discourses on social media platforms.
Therefore, we propose an early reflection on and incorporation of the
range of tools, metrics, and methods commonly referred to as Social Media
Analytics (Zeng, Chen, Lusch, & Li, 2010) in the methodological frame-
work of a study, to provide a bird’s-eye view of the communicative environ-
ment and to identify the patterns and objects that are more or less visible
because of the social media logics embedded into the platform. Metrics dif-
fer by platform studied, and must be systematically and critically considered.
For Twitter-centric studies, Bruns and Stieglitz (2012, 2014) have proposed
a range of metrics and patterns that are commonly of interest. However,
such metrics should not be treated merely as objective statistical measures of
engagement, participation, or interaction. Rather, any identification of what
is visible on a platform should be accompanied by follow-up reflections on
how and why it is visible, the discursive functions this visibility achieves, and
the materialities giving rise to its visibility and thus to its accumulation of dis-
cursive power. Finally, each research project has its own questions and inter-
ests, which require researchers to adapt their methodological choices to the
challenges and limitations of studying a certain platform—such as issues of
data quality, topic discovery and delineation, and data processing (Stieglitz,
Mirbabaie, Ross, & Neuberger, 2018)—rather than apply a one-size-fits-all,
off-the-shelf analytics solution.
Social Network Analytics

Following on from the initial exploratory insights drawn from the social
media analytics phase, another important level of analytical possibilities cre-
ated by social media data can be broadly conceptualized as “networkedness.”
Social actors, media objects, platform features, and interactions among them
created as a result of platform affordances can be conceived as building blocks
of a network of interactions and articulatory practices. Depending on the
platform, this could be a network of accounts following a certain page, com-
menting on a subreddit, subscribing to a YouTube channel, @mentioning and
retweeting other accounts, and many more possibilities. These collective dis-
cursive actions point to certain common features among actors and elements
in a social environment (McPherson, Smith-Lovin, & Cook, 2001). Someone
interested in videogames, for instance, may be more likely to interact with
actors and content that reflect this interest. The same is true for other dis-
courses and discursive practices.
Therefore, any collective discursive practice can exhibit homophilic ten-
dencies that lead to the clustering of related elements. The detection and
interpretation of such clusters provide the researcher with crucial informa-
tion about the discourses and practices present and prevalent in a large vol-
ume of data. The second phase in our methodology involves employing the
range of tools, techniques, and methods known as Social Network Analysis
(SNA). Although one can generate similar findings through a manual reading
and coding of the dataset, this would be very time consuming for large-scale
datasets. Instead, as we show below, SNA can provide a faster way of identi-
fying the various discursive positions in a dataset.
In the simplest of terms, community detection algorithms follow the
assumption discussed above and cluster different elements together if there
is heightened interaction among them (Fortunato, 2010). These clusters can
act as primary indicators for the investigation of the discourses and discursive
strategies present in the data. However, we emphasize again that these tools
are not a one-size-fits-all solution, because they rely only on mathematical and
statistical techniques in grouping and clustering elements together. A cluster
of elements identified by an algorithm may make complete statistical sense,
but fail to reveal any meaningful underlying discursive structures; the clusters
identified by an algorithm always also require a qualitative, context-aware
examination and interpretation. This is why we argue for a discourse-oriented
framework in understanding the results of a social network analysis. Below,
we show how different network structures can reveal valuable information
about the discursive structures and strategies present in a study.
Textual Analysis
In the third phase of the methodology, we turn to the textual representations
of discourses in the dataset. As argued above, an examination of a collective
social practice cannot be considered satisfactory unless it accounts for the
context and the broader discursive environment. This generally requires some
form of in-depth, qualitative, context-aware analysis.
The first two stages of the methodological procedure discussed here
point to a series of objects, discourses, and collective discursive formations
that require further qualitative investigation. For instance, upon discover-
ing a cluster of accounts with a high level of insider retweeting and a low
level of outsider interactions, we are faced with a discursive formation created
through the articulatory practice of retweets. A qualitative review of the key
accounts in the cluster (where “key” may be defined by the metrics gener-
ated through generic social media analytics) often reveals information about
the shared discursive interests of the accounts in the cluster. This can be an
entry point to the data selection and sampling for a discourse-analytical study,
enabling the researcher to select purposive samples of tweets from the key
accounts in the cluster, or a stratified random selection from all the accounts
in the cluster, depending on the research question. However, this might still
pose significant difficulties, especially for large datasets or studies interested in
everyday practices and discourses.
In bridging the gap between large-scale datasets and the smaller sam-
ples required for an in-depth discourse-analytical study, scholars have noted
the appropriateness of corpus linguistics approaches (Baker, 2012; Baker &
Levon, 2015; Evans, 2014; Wiedemann, 2013). Techniques and approaches
for keyword analysis have proven useful for (critical) discourse-analytical
studies; they can be used to identify salient themes, topics, and discursive
strategies in a corpus of systematically chosen texts (Baker, 2004). Going
back to the example of a cluster of densely connected accounts identified in
the network analysis stage, all the posts/tweets/comments by the accounts
in that cluster could be treated as a corpus of texts written around a shared
discourse. Depending on the aims of the study, this corpus could be com-
pared to a larger, generic reference corpus in order to identify its most
distinct keywords, salient themes, and central topics. This positions the
researcher much more comfortably for further in-depth discourse-analytical
investigation.
Practical Applications for Cyclical Mixed-Methods Social Media

Analysis
Having discussed the broader methodological considerations, we will now
show how these methods can be used to provide context-aware, discourse-ori-
ented insights into social media data. It is beyond the scope of this chapter
to report the findings of all stages of these studies. Therefore, we focus only
on elements of the two case studies that elucidate the preceding argument.
Case Studies: Political Discussions in the Australian Twittersphere

As noted, the methodological reflections we have introduced here can be
adapted to account for different theoretical frameworks and theories of
discourse. In our two case studies, we draw from the discourse theory of
Laclau and Mouffe (2001) to show how the different metrics and analyti-
cal steps could be examined through the lens of discourse theory, and how
a discourse-analytical study could benefit from the incorporation of these
tools and metrics to enrich its findings, account for the materialities of the
platform, and manage the sheer volume of data acquired from social media.
Although we use Twitter data in these examples, the proposed methodology
could be used for data collected from other platforms as well, with modifica-
tions as needed. Before presenting our findings, we briefly introduce the two
case studies.
#RoboDebt: An Algorithmic Scandal

In 2016, the Australian government started using an automated data-match-
ing algorithm to discover overpayments made to individuals receiving wel-
fare assistance from Centrelink, the welfare program run by the Australian
Department of Human Services. The algorithm compared an individual’s
annual employment records from the Australian Tax Office with the payments
received from Centrelink; if discrepancies were found, the system automati-
cally issued a debt notice letter, stating the amount owed to the government
and giving a deadline to repay the money or prove that the person did not
owe the debt identified. It did not take long before numerous reports of flaws
in the system, incorrect debt notices, and problems in the process found their
way into the media. On Twitter and elsewhere, activists started referring to
the issue with the term (and hashtag) “RoboDebt,” and formed a campaign
under the slogan “Not My Debt” to protest unfair treatment under this algo-
rithmic data-matching regime. The scale of reports, campaigning activities,
and reports of individuals receiving debt notices of many thousands of dollars
led to two external investigations, both of which found flaws in how the algo-
rithm was implemented and the complaints were handled.
We collected tweets discussing this campaign between December 2016
and May 2017, by tracking the keywords and hashtags Centrelink/Centerlink,
Tudge (the Human Services Minister at the time), Robodebt/#robodebt,
and NotMyDebt/not my debt/#notmydebt. This resulted in a dataset of
about half a million tweets. The #RoboDebt case study serves as a useful
indicator of the different voices, discourses, and actors that may be involved
in a social media discussion about a government scandal. It shows how com-
munities and other stakeholders strategically draw from different symbolic
resources and interdiscursive references to resist a social injustice. Further, it
shows how the materialities of a social media platform and its affordances help
in the creation of virtual discourse communities.
Section 18C of the Racial Discrimination Act

For our second case study, we focus on a long-standing and divisive political
debate in the Australian public sphere. Over the past decade, increasing ten-
sions related to freedom of speech in Australia have often revolved around
Section 18C of the Racial Discrimination Act, which makes it unlawful to
publicly discriminate against individuals or groups based on their ethnic-
ity, race, religion, or other personal attributes if to do so is reasonably likely
to “offend, insult, humiliate, or intimidate” them. In particular, the words
“offend” and “insult” have been the focus of much controversy. Critics of
the law argue that these words are subjective and open to interpretation.
A number of Australian politicians have called for either a replacement of
these terms with the stronger word “harass” or a complete repeal of 18C. In
2017, the Australian Senate voted down any proposals for change, and the
section has remained intact. Our dataset of tweets discussing Section 18C,
posted between August 2016 and May 2017, contains just under 200,000
tweets containing keywords and hashtags 18C, S18C, Section 18C, Racial
Discrimination Act, or RDA.
Agonism and Antagonism: Distinguishing Discursive Patterns in Social

Media Debates
A complete discussion of the history, context, and findings of these two case
studies is not possible here. Instead, we review elements of the social network
analysis stage, and demonstrate how discourse theory guides making sense
of the interaction networks, informs data sampling, and underpins further
qualitative investigation. In particular, building on Mouffe (2013), we focus
on evaluating the agonistic and antagonistic qualities of the discourses we
observe; in this context, agonistic represents “a struggle between adversar-
ies,” whereas antagonistic is “a struggle between enemies” (Mouffe, 2013,
p. 17). Laclau and Mouffe (2001) conceptualize discursive struggle as an
ineradicable and necessary condition of democracy. In such an understanding,
the goal of a democratic project is not to reach to consensus, but to channel
antagonisms in a way that they are transformed from a struggle between ene-
mies—the goal of which is to eliminate the enemy—to an acknowledgment
of differences between adversaries and working with the adversary toward a
common goal (Mouffe, 1999).
In both cases, the retweet networks could be understood as networks
of discursive amplification. Although a retweet does not necessarily mean an
endorsement of a message, it is an indicator that a user deems something
worthy of being circulated and seen by her or his followers. Thus, the net-
work clusters of accounts created through retweeting as an articulatory prac-
tice point to a collective perception of the sorts of messages that these users
believe are in need of circulation.
In our two examples, the retweet patterns (Figures 10.1 and 10.2)
point to quite different network structures. The network of retweets in the
#RoboDebt case study forms a star-shaped constellation of users retweeting a
small number of key accounts in the discussion. In this core-periphery struc-
ture, almost all users discussing the issue retweet certain accounts whom they
presumably consider influencers in the debate. These activist accounts, which
can be understood as crowd-sourced elites (Papacharissi, 2015), are those
Figure 10.1. Retweet network of the #RoboDebt study
Figure 10.2. Retweet network of the 18C study

whose messages are perceived by Twitter users to be so significant that they

should be seen by as many others as possible. Consequently, they are posi-
tioned in the center of the network. The lack of any other clusters in the net-
work also shows that there might not be any competing discourses and actors
regarding this issue. In other words, in curating and channeling the discus-
sion about RoboDebt, there appears to be a consensus on the key participants
and whose voices need to be amplified to greater visibility. This network for-
mation resembles the Broadcast Network in Smith, Rainie, Himelboim, and
Shneiderman’s typology of communication patterns on Twitter (2014). As
we discuss below, this could be an indicator of an agonistic space, created
through retweets as one of the key affordances of Twitter, in which users with
competing ideologies and discourses have formed a discursive alliance that at
least temporarily transcends latent antagonisms for the sake of expressing a
shared view or achieving a common objective.
In contrast, the retweet network of the 18C case study forms quite a
different structure. Three distinct clusters of users exhibit a higher propensity
to retweet fellow members of their cluster than to retweet others outside of
the cluster. Of these three clusters, one is comparatively less connected to the
other two. This structure sits somewhere between Smith et al.’s Polarized
Crowds and Community Clusters network types (2014), without entirely
resembling either of these ideal types. Following the points made in the pre-
vious section, in such cases it is usually necessary to qualitatively examine the
core accounts in each cluster and thereby contextualize these accounts in
order to make inferences about the discourses shared and worldviews repre-
sented by the accounts in each cluster.
A qualitative examination of the three clusters in this network, by review-
ing the profile information and tweets of key accounts, reveals three distinct
groups of users: those generally opposing changes to Section 18C, Indigenous
rights activists who have specific reasons to oppose changes, and those calling
for a repeal of the section. The clusters of 18C supporters and Indigenous
rights activists are tightly interconnected and therefore comparatively close to
each other, whereas the repeal cluster has fewer connections to the other two.
This can be explained by the close attitudinal alignment between the two pro-
18C clusters: over the course of the debate, Indigenous activist organizations
in Australia campaigned actively against making any changes to Section 18C,
arguing that the proposed changes could put Aboriginal and Torres Strait
Islander communities at an increased risk of racial vilification and racism.
The structure of retweet networks in this case reveals two interesting
discursive practices. First, the polarized network structure between the pro-
and anti-18C actors points to the presence of antagonistic discourses and
discourse communities. This was predictable, but it is nonetheless important

that the empirical evidence supports our working hypothesis on this point.
Second, and perhaps more interesting, the distinct yet closely related clus-
ters of Indigenous rights activists and progressive users resisting changes to
18C indicate a discursive alliance and an agonistic discourse, formed around
the notions of social justice, anti-racism, and equality as its nodal points,
that envelops both communities; at the same time, however, the distinction
between these two clusters on the pro-18C side shows that these discursive
allies maintain their own identities and discourses through drawing from dif-
ferent symbolic resources, even as they form a discursive alliance against a
common antagonistic opponent. In this case, the affordance of retweeting
provided by Twitter helps in the formation of a shared agonistic discursive
practice across two otherwise distinct communities.
Using the example of the retweet networks, we have shown how network
analysis can lead a discourse analyst to form a working hypothesis about the
range of discourses and discursive practices present in the data under investi-
gation. We have also shown how awareness of context and discourse theory
can guide a network analyst in connecting network structures to socio-po-
litical findings. However, the analysis should not necessarily conclude at this
early interpretive stage; especially for a discourse analyst, the linguistic and
textual representations of discourse remain central to the study. From this
perspective, in addition to identifying discursive formations and communities,
the findings of the network analysis stage should inform further data sampling
considerations. For instance, in the case of the Section 18C debate, two major
communities of pro- and anti-18C users were identified through network
analysis. The discourse analyst can construct a purposeful sample of tweets
posted in each cluster to serve as the basis for further qualitative investigation.
In approaching this sampling, although one may focus on tweets posted
by the core accounts in each cluster (to study the discourse of key activists
and influencers), our research interest in the present project was the collective
discourse among members of a discursive cluster. It is essentially impossible to
conduct a qualitative discourse-analytical study of the hundreds of thousands
of tweets posted by the members of each cluster. However, corpus linguistics
tools can help identify the themes, topics, and keywords salient in the dis-
course of each cluster community in comparison with the broader discussions
around Section 18C. We used the tweets posted by each community as the
sample corpus, and all tweets in the dataset as the reference corpus. This tech-
nique assigns a rank to each keyword in the corpus, based on its frequency
and statistical significance (i.e., on how unusually frequent the keyword is
in the sample corpus as compared to the reference corpus). This enables us
to identify features that are salient in the discourse of a certain community

(e.g. because they express pro- or anti-18C sentiment), including the themes,
topics, and discursive strategies, the symbolic resources, and other features.
(Such comparative analysis is impossible in a single-cluster dataset as encoun-
tered in the #RoboDebt case; still, the discourse of the anti-RoboDebt com-
munity on Twitter could perhaps be compared to the pro-RoboDebt rhetoric
of government statements and interviews.)
The keyword analysis of the corpus of tweets from each community
(Table 10.1) points us to the presence of two competing, antagonistic dis-
courses. The pro-18C community of users frames the discussion by using
interdiscursive references to the discourse of hate speech, White supremacy,
and racism, arguing that the proposed changes pushed by liberal politicians
and supported by right-wing groups in Australia will eventually lead to an
increase in hate speech. Such references are almost absent from the discourse
of the anti-18C community, which foregrounds discourses of free speech
and criticisms of Islam and political correctness, arguing that Section 18C
must either be reformed or repealed in order to ensure freedom of speech
in Australia, including the freedom to criticize Islam. Additionally, as Reisigl
and Wodak (2005) argued, discursive strategies of nomination (how different
Table 10.1. Keywords in the discourse of pro- and anti-18C communities
Anti-18C Discourse Keywords Pro-18C Discourse Keywords

Rank Keyword Rank Keyword
5 reform 6 changes
6 petition 7 right
7 must 17 libs
8 islam 18 white
14 sign 25 liberals
15 should 32 rwnjs
16 repeal 33 rw
31 left 37 hate
35 removed 38 liberal
41 pc 41 agenda
52 abolished 45 rwnj
53 rid 49 racists
66 dismiss 50 bigots
Ranking of distinct keywords in each discourse community. Shared and generic terms (e.g. prepositions,
articles, etc.) excluded from each list.
actors, events, and so forth are named and referred to) and predication (the
qualities attributed to different actors, events, and so forth) are often pres-
ent in antagonistic discourses, which focus on an us/them dichotomy. Such
strategies are frequently present in the discourses of the clusters we identi-
fied in the retweet network. In the discourse of both communities, there
are frequent uses of diminutive pejoratives in reference to the Other: anti-
18C discourses refer to the Other with terms such as “SJW” (“Social Justice
Warrior”) or “lefty snowflake,” whereas pro-18C users employ pejoratives
such as “free speech warrior”, “RWNJ” (“Right Wing Nut Job”), or “old
white male.”
It is beyond the scope of this chapter to provide a full discourse-analyt-
ical account of the findings of these two case studies; detailed results of this
research will be presented in other contexts. Instead, our intent is to outline
the broad methodological considerations in undertaking such mixed-meth-
ods investigations. Our examples show how methods such as social media
analytics and social network analysis provide researchers with the tools to
move from the large, noisy datasets obtained from social media platforms
to a small, objectively selected sample of texts that is ready for further qual-
itative analysis. They also show how well-established discourse theories can
inform working hypotheses about the bird’s-eye participation and interaction
patterns obtained through quantitative approaches such as social media ana-
lytics and social network analysis—hypotheses that can be further evaluated
through purposeful qualitative analysis.
Enlisting Computational Methods in Doing Discourse Analysis

Not least because of the potential size of the big social data collections they
make available, social media continue to present a substantial challenge to
researchers in the humanities and social sciences, and especially to scholars
from traditionally more qualitatively focused disciplines. Recent years have
seen a “computational turn” (Berry, 2011) toward new research practices that
are sometimes categorized under umbrella terms such as “digital humanities”
or “computational communication studies,” but it is critical that such devel-
opments are not misunderstood as implying a wholesale move from quali-
tative to quantitative approaches; purely quantitative research methods are
valuable for particular research questions, but tend to produce aggregate and
bird’s-eye perspectives that identify broad patterns yet lack equally import-
ant insights into the finer detail. Conversely, purely qualitative investigations
reveal details, but are often unable to contextualize the findings against the
backdrop of larger communicative spaces.
Rather, as we argue, the combination of large-scale quantitative and

fine-grained qualitative methods in genuine mixed-methods research designs
presents a particularly fruitful path toward methodological advancement. In
the abstract, this is not a new observation; what we have provided is a struc-
tured overview of the concrete tools and practical steps involved in working
through a research agenda that interweaves both quantitative and qualitative
elements. Our brief discussion of how we have applied this research agenda to
two political debates on Twitter in Australia, and of how we have operation-
alized this research in pursuit of questions emerging from discourse theory,
demonstrates the use and utility of this mixed-methods approach in everyday
research practice.
The methodological frameworks and tools for social media analytics,
social network analysis, and textual analysis presented here are not the only
such methods that could have been applied to the two case studies, nor nec-
essarily the most sophisticated; methodological innovations since the emer-
gence of big social data have produced a vast array of approaches and tools
that can be brought to bear on datasets representing social media and other
forms of communication to the point that any attempt to comprehensively
catalog them would be a fool’s errand. Researchers should always seek to
develop and maintain their methodological literacies, but happily even the
fairly basic methods we have employed produce valuable results when they are
sensibly combined and sequenced. It is less important to employ the very lat-
est and most complex methods than to draw on methodological approaches
that are relevant to the research questions being asked.
Although applicable to a broader range of domains, our observations are
directed most immediately to discourse analysts, who in the context of social
media data continue to struggle especially with the purposeful selection of
appropriate samples. As in the case of the RoboDebt and 18C examples, the
pathway from large-scale datasets through the application of standard quan-
titative analytical methods to the selection of meaningful samples for further
discourse analysis should help address this challenge, and in doing so support
and enable new discourse-analytical studies of communicative phenomena in
social media environments.
References
Baker, P. (2004). Querying keywords: Questions of difference, frequency, and sense in
keywords analysis. Journal of English Linguistics, 32(4), 346–359.
Baker, P. (2012). Acceptable bias? Using corpus linguistics methods with critical discourse
analysis. Critical Discourse Studies, 9(3), 247–256.
Baker, P., & Levon, E. (2015). Picking the right cherries? A comparison of corpus-based and
qualitative analyses of news articles about masculinity. Discourse & Communication,
9(2), 221–236.
Berry, D. M. (2011). The computational turn: Thinking about the digital humanities.
Culture Machine, 12. Retrieved from http://www.culturemachine.net/index.php/
cm/article/view/440
boyd, d., & Crawford, K. (2012). Critical questions for big data: Provocations for a cul-
tural, technological, and scholarly phenomenon. Information, Communication &
Society, 15(5), 662–679.
Bruns, A. (2019) Are filter bubbles real? Cambridge, UK: Polity.
Bruns, A., & Stieglitz, S. (2012). Quantitative approaches to comparing communication
patterns on Twitter. Journal of Technology in Human Services, 30(3–4), 160–185.
Bruns, A., & Stieglitz, S. (2014). Twitter data: What do they represent? it—Information
Technology, 56(5), 240–245.
Burgess, J., & Bruns, A. (2015). Easy data, hard data: The politics and pragmatics of
Twitter research after the computational turn. In G. Langlois, J. Redden, & G. Elmer
(Eds.), Compromised data: From social media to big data (pp. 93–111). London,
UK: Bloomsbury.
Carpentier, N. (2017). The discursive-material knot: Cyprus in conflict and community
media participation. New York, NY: Peter Lang.
Evans, M. S. (2014). A computational approach to qualitative analysis in large textual
datasets. PLoS ONE, 9(2), e87908.
Fortunato, S. (2010). Community detection in graphs. Physics Reports, 486(3–5), 75–174.
Foucault, M. (1971). Orders of discourse. Social Science Information, 10(2), 7–30.
Halavais, A. (2015). Bigger sociological imaginations: Framing big social data theory and
methods. Information, Communication & Society, 18(5), 583–594.
KhosraviNik, M., & Esposito, E. (2018). Online hate, digital discourse and cri-
tique: Exploring digitally-mediated discursive practices of gender-based hostility. Lodz
Papers in Pragmatics, 14(1), 45–68.
Laclau, E., & Mouffe, C. (2001). Hegemony and socialist strategy: Towards a radical dem-
ocratic politics (2nd ed.). London, UK: Verso.
Manovich, L. (2012). Trending: The promises and the challenges of big social data. In
debates in the digital humanities. Minneapolis, MN: University of Minnesota Press.
McPherson, M., Smith-Lovin, L., & Cook, J. M. (2001). Birds of a feather: Homophily
in social networks. Annual Review of Sociology, 27(1), 415–444.
Mouffe, C. (1999). Deliberative democracy or agonistic pluralism?. Social Research, 66(3),
745–758.
Mouffe, C. (2013). Agonistics: Thinking the world politically. London, UK: Verso.
Papacharissi, Z. (2015). Affective publics: Sentiment, technology, and politics. New York,
NY: Oxford University Press.
Reisigl, M., & Wodak, R. (2005). Discourse and discrimination: Rhetorics of racism and
antisemitism. London, UK: Routledge.
Smith, M. A., Rainie, L., Himelboim, I., & Shneiderman, B. (2014). Mapping Twitter
topic networks: From polarized crowds to community clusters. Pew Research Center,
20, 1–56.
Stieglitz, S., Mirbabaie, M., Ross, B., & Neuberger, C. (2018). Social media analytics—
Challenges in topic discovery, data collection, and data preparation. International
Journal of Information Management, 39, 156–168.
van Dijck, J., & Poell, T. (2013). Understanding social media logic. Media and
Communication, 1(1), 2–14.
Wiedemann, G. (2013). Opening up to big data: Computer-assisted analysis of textual
data in social sciences. Historical Social Research/Historische Sozialforschung, 14(2),
332–357.
Zeng, D., Chen, H., Lusch, R., & Li, S.-H. (2010). Social media analytics and intelli-
gence. IEEE Intelligent Systems, 25(6), 13–16.
11.
The Hyperperception Model: How
Observing Others on Social
Media Can Affect People in Close
Relationships
Erin L. Spottswood and Christopher J. Carpenter
After Lydia uploads a photo of her lunch to Instagram, she scrolls through
her feed and sees her boyfriend Paul has posted a photo featuring himself and
some colleagues getting coffee. He is standing quite close to a woman she
doesn’t recognize; this unknown woman left a heart on his post and com-
mented “Thanks for the java!” Disquieted, Lydia clicks on Paul’s profile and
notices Miss Java in a few recent posts, and that she has liked or commented
on a lot of Paul’s posts. Lydia clicks on Miss Java’s profile only to realize it is
private. Nagging suspicions arise: Who is Miss Java? Why didn’t Paul mention
her? Lydia knows she is just being paranoid … right?
It had been several weeks since Maeve returned to college after break,
and her dad Mark hasn’t heard from her. Sure, he could reach out to her but
he doesn’t want to intrude. He looks at Maeve’s Facebook Timeline to get a
sense of what she’s been up to, and sees recent posts about New Year’s Eve,
hiking with friends, and hosting a dinner party—all of which feature a young
man Mark does not recognize. Mark noticed that Mr. Mystery had liked all
these recent posts, and commented on the New Year’s Eve post “Clear eyes,
full hearts, can’t lose!” What did that mean?
These examples illustrate how Instagram, Facebook, Twitter, and other
social media make other people’s interactions easily accessible, reviewable,
and open to interpretation. Those observations can be innocuous (Paul is
just getting coffee with colleagues; Maeve is just having fun before classes
start), but they could also be interpreted as suspicious (my boyfriend’s new
colleague has a crush on him; a stranger is pursuing my daughter) depending
178 Spottswood & Carpenter
on the interaction norms people associate with the technology, the relation-
ship between the observer and the observed, and how much information the
observer has to contextualize the interactions.
The goal of this chapter is to describe a new theoretical model, the hyper-
perception model (Carpenter & Spottswood, 2019) as well as showcase a prelim-
inary test of some of the model’s key components. This new theoretical model
aims to predict, explain, and describe how and why users sometimes develop
inflated impressions of other people’s interactions and relationships according
to what they observe and interpret on social networking sites and/or social
media applications (henceforth we will use the term “social media” when
referring to both sites and applications). This chapter describes the hyperper-
ception model’s predecessor, the hyperpersonal model (Walther, 1996), explain
why new models may be needed to account for changes in the ways people use
communication technology today, present the hyperperception model and its
purpose, showcase a preliminary test of some of the hyperperception model’s
principal components, and conclude with suggestions for future research.
The Hyperpersonal Model
The hyperperception model is largely inspired by Walther’s (1996) hyperper-
sonal model, which posits that computer-mediated communication (CMC)
channels can foster interactions and relations that can surpass the level of
affection and emotion experienced in analogous face-to-face (FtF) contexts
so long as its four components are in place. The first component focuses on
the CMC channels that can both constrain the amount of nonverbal cues that
are typically available during FtF interaction and enable interaction partners
more time to reflect, compose, send, and interpret each other’s messages. So
long as users perceive that they can take advantage of these aspects of a com-
munication technology, they will be able to engage in hyperpersonal inter-
action via that channel. The second component of the hyperpersonal model
focuses on senders who take advantage of CMC channels to carefully craft a
selective presentation of themselves that heightens their social attractiveness.
The hyperpersonal model focuses on channels that primarily support verbal
communication (e.g., email or electronic bulletin boards) and give senders the
ability to be especially selective about their online presentation because textual
or verbal information is more “malleable” and “subject to self-censorship”
than nonverbal information (Walther, 1996, p. 20). The third component
of the hyperpersonal model focuses on receivers who make over-attributions
about a sender’s social qualities. Over-attributions occur when receivers focus
on the sender’s disclosures that appeal to them and pay less attention to
The Hyperperception Model 179
disclosures that are less appealing. This will paint an especially socially attrac-
tive picture of the sender in the receiver’s mind. The fourth component of
the hyperpersonal model presents a feedback loop wherein receivers who come
upon selectively self-presented senders idealize those senders and respond in
ways that support the senders’ presentation. This may lead senders to start
communicating and behaving in ways that further confirm the receivers’ ide-
alizations, that is, result in behavioral confirmation.
The hyperpersonal model is incredibly robust as is the research that
has tested its components across a variety of CMC contexts. For example,
research has shown that people take advantage of CMC (Walther, 2007)
and social media channels (Gonzales & Hancock, 2011) to project selective
self-presentations to other users of these channels. Research has also shown
that receivers sometimes make over-attributions about their interaction part-
ner’s social qualities according to what the sender self-presents as well as their
own interpersonal motivations (Anderson & Emmers-Sommer, 2006; Brody,
2013). McEwan and Zanolla (2013) found that people who interact exclu-
sively in an online environment are sometimes disappointed when they finally
meet their interaction partners face-to-face; this disappointment is posited to
be consistent with the feedback loop component because it seemed the inter-
action partners were less able to engage in the same behavioral confirmation
during FtF interaction as they could when they were interacting online. This
is just a sliver of the research that is consistent with the hyperpersonal mod-
el’s components; however, technological advances warrant a new look at the
suitability of this model.
In the 1990s people could observe others interacting in online chatrooms
or online communities (e.g., listservs, message boards, etc.), but because these
CMC channels were primarily text-based and did not necessarily require peo-
ple to disclose identifiable information digitally or normatively, verifying who
(e.g., observed senders) was talking to whom (e.g., observed receivers) was
difficult if not impossible. However, today’s social media channels allow peo-
ple to surveil or observe people interacting in an environment in which their
interactions are typically associated with people who have verifiable offline
identities. Social media channels and social norms also make it possible for
observers to access others’ conversations so long as the conversations persist
on a given website or mobile application.
The Hyperperception Model
Just as Walther’s (1996) hyperpersonal model helps explain and predict what
happens when people meet, interact, and develop a relationship in a primarily
text-based CMC environment, the hyperperception model extends key aspects

of the hyperpersonal model to illuminate why, when, and how people develop
inflated impressions of others’ interactions and relationships based on what
they see on a social media channel. Hyperperception occurs when a third-
party observer perceives the interactions and relationship between two people
on a social media channel as more intense than those being observed would
report. Intensity refers to a class of relational variables such as intimacy, close-
ness, similarity, tie strength, and emotional involvement.
The hyperperception model is comprised of four components that explain
how and why social media channels and social media users can at times lead
an observer to hyperperceive other people’s interactions and relationships.
The hyperperception model’s components parallel the original hyperpersonal
model’s (Walther, 1996) channel, sender, receiver, and feedback loop com-
ponents. However, the difference is that instead of focusing solely on the
people interacting via CMC, the hyperperception model also addresses the
affordances of social media channels and the observer of the interactions. The
hyperperception model posits that given unique aspects of social media chan-
nels, users are better able to observe other people’s interactions, and that in
turn affects how observers perceive the interactions and relationships of those
who are being observed on social media.
The hyperperception model posits that there are times when a person is
especially motivated to observe other people’s interactions and relationships
displayed on a social media channel. According to this model, the observer is
the person using social media channels to appraise another person’s interac-
tions and relationships. Generally, two types of individuals are observed. The
observed sender is someone the observer knows and is directly connected to
on a social media channel. The observed receiver is the person with whom the
observed sender interacts and is someone the observer does not know well or
at all. The observed sender and observed receiver roles are not fixed, especially
when the two people being observed on a social media channel interact on
that (and possibly other) channels, but these labels make our description easier
to follow and apply to interpersonal phenomena discussed here and elsewhere.
Before delving more into the observer’s motivations, it is crucial to explicate
how aspects of social media channels can make hyperperceptions not only pos-
sible but also even more likely to occur in the social media environment than
when a sender and receiver are observed interacting on or via other channels.
The Channel
The first component in the hyperperception model is the social media chan-
nel being used to observe others’ interactions. Social media affordances are
aspects of the channel, the person, and the social context that affect how
people use and make sense of what they see on a social media channel (boyd,
2010; Evans, Pearce, Vitak, & Treem, 2016). The affordances that increase
the likelihood of hyperperception effects are accessibility (the ability to access
other’s interactions on a social media channel), persistence (the length of time
an interaction remains visible on a social media channel), and association (the
extent to which an interaction can be linked to an identifiable person on and
off a social media channel).
First, the observed sender-receiver interactions must be accessible to the
observer. For example, private or direct messages cannot trigger hyperper-
ceptions because the observer cannot access those interactions. However,
if an observed sender posts a status update to which the observed receiver
responds, this relatively publicly accessible record of interactions is visible and
capable of being appraised by the observer.
Second, the observed sender-receiver interactions must be persistent such
that the observer can view the dyad’s recent and past interactions. Moreover,
repetitive observation may be especially likely to produce hyperperception
effects if the social media channel being used is endowed with norms that
encourage users to post positive and socially attractive content in their posts
and interactions with each other (Utz, 2015).
Third, the observed sender-receiver interactions must be associable or
linkable such that the observers can identify the people they are observ-
ing and appraise the interactions and relationship. Examples include a wife
(observer) wondering whether her husband (observed sender) is flirting with
a coworker (observed receiver), or a father (observer) who wants to know
whether his daughter (observed sender) is developing a romantic relationship
with someone (observed receiver). Before CMC, especially social media, peo-
ple did not have easy access to such records of people they know interacting
with other people they know less well or not know at all. But contemporary
CMC offers access to these records and thereby helps make hyperperception
possible. These aspects of the channel are required for hyperperception effects
because the channel must give the observer something to observe.
The Observed Sender
The second component of the hyperperception model is the observed sender
and is two-pronged. First, the observed sender must be someone that the
observer is motivated to observe. Of the multitude of people who the observer
is connected to via social media, the observer is likely interested in keeping
tabs on only a few people (e.g., a romantic partner, close friend, family mem-
ber, a potential romantic partner or friend). We note that observing others is
not necessarily motivated by insecurity; people generally like to know about

others’ social lives (Dunbar, 2004). Many people in romantic relationships
have a desire to observe their romantic partners’ interactions with other peo-
ple to assess the fidelity of their relationship (Buss, 1988).
In addition, the observer must perceive that the observed sender is sin-
gling out one or maybe a few specific people (the observed receivers) via their
posts and interactions on a social media channel. In the context of a romantic
relationship, hyperperception requires that the observer believes his or her
partner (the observed sender) may be romantically interested in at least one
other person (an observed receiver) or vice versa. Again, the observer may be
wrong; the partner may just be following the posting and interaction norms
of the social media channel being used (Spottswood & Hancock, 2016; Utz,
2015). The observer’s partner may have no desire to escalate a relationship
with the observer. But when the observer’s partner’s (the observed sender)
social media interactions with another person (the observed receiver) make it
seem as though the latter two individuals are socially attracted to each other
(their interactions are frequent, cover a wide range of topics, sometimes delve
deep into personal topics), the observer may make (inaccurate) assumptions
about the sender and receiver’s interactional and relational intimacy. This
is when the observer makes a hyperperception about the observed sender’s
interactions with another person (observed receiver). Hyperperceptions are
also more likely if the observer has constrained access to the observed receiver
on and off social media.
The Observed Receiver

The third component focuses on the observed receiver(s). When an observer
is not able to view the observed receiver’s interactions with others besides the
sender on a social media channel or via other digital and corporeal channels,
the observer may have difficulty determining whether the perceived inten-
sity of the observed sender and observed receiver’s social media interactions
is attributable to a heightened closeness between the pair or whether the
observed receiver typically acts that way toward others. Constrained access is
more likely the less an observer knows the observed receiver on- and offline
given that people are less inclined to connect to strangers on social media
channels (Patil, 2012). Not being connected on social media will make it
difficult for the observer to access and assess the observed receiver’s interac-
tions and relationships with others besides the observed sender. The less an
observer is able to compare the observed sender-receiver interactions with
the observed receiver’s interactions with other people besides the observed
sender, the more likely it is that the observer may perceive that the observed
sender-receiver relationship is unique to the observed sender and observed

receiver. This is compounded by the fact that CMC and social media channels
have fewer nonverbal immediacy cues (e.g., proximity, body direction, and
touch) that often indicate how close two people really are or want to be. Were
the observer to see the dyad interacting offline, many cues of intimacy would
likely be available and the observer would be less likely to make hyperpercep-
tions about the sender-receiver interactions and relationship.
Moreover, Walther (1996) noted that in many CMC environments, peo-
ple produce an unusually positive self-presentation. This is also the case on
social media channels (Fox & Vendemia, 2016), so what little of the observed
receiver the observers can access may look particularly attractive—especially if
the observed receivers selectively self-present in ways that make them look as
attractive, intelligent, and interesting as possible (Hogan, 2010). In the case
of romantic relationships, as discussed below, these factors make the observed
receiver more likely to seem like a potential romantic rival and thus are likely
to lead to hyperperception. As such, constrained access to the observed
receiver’s interactions on and off a social media channel increases the likeli-
hood that an observer will make a hyperperception about the observed sender
and observed receiver’s relationship.
The Feedback Loop
The fourth component posits that if the three preceding components co-oc-
cur to a considerable degree, the observer may sense an escalation in the
intensity of the interactions and/or relationship between the observed sender
and observed receiver. Specifically, if the observer perceives evidence of partic-
ularly intense interactions between the observed sender and observed receiver
on a social media channel, the observer may be motivated to observe past and
future interactions between the pair. These additional and repeated observa-
tions may provide the observer with further self-proclaimed “evidence” of an
increasingly intense relationship, which in turn might motivate further sur-
veillance. This feedback loop is especially likely if the observed sender-receiver
interactions persist on the social media channel. Such a feedback loop has been
described in the context of Facebook-related jealousy (Muise, Christofides, &
Desmarais, 2009) and is included in the hyperperception model to describe
how hyperperception can build over time.
Preliminary Test of the Hyperperception Model

As a preliminary test of our model, we conducted an online survey where
participants were given many opportunities to discuss what happened when
their romantic partners seemed to be hyperperceiving the participants’ inter-

actions with other social media users. In other words, our participants were
the observed senders in the hyperperception model, who reported how
their partner (i.e., the observer) was using a social media channel to observe
them interact with others (i.e., observed receivers). The participants’ survey
responses were examined for illustrative evidence of some of the processes
described by the hyperperception model as well as for potential new avenues
for research.
Participants were recruited via a mass email to the students of a medi-
um-sized, Midwestern US public university. They were offered a chance to
win $30 in exchange for their participation. They had to either be currently
in a romantic relationship (N = 134) or have formerly been in a roman-
tic relationship (N = 33). Of the participants currently in a romantic rela-
tionship, 15 reported seeing their current partner once a month or less, 36
reported every few weeks or once a week, and 81 reported a few times a week
or more. Of those who responded to demographic questions, 48.4% iden-
tified as female, 48.4% as male, and 1.2% as fluid; the mean age was 22.33
(SD = 3.69). Of those currently in a relationship, the average duration was
4.04 years (SD = 4.86). Of those describing former relationships, the average
duration of those relationships was 1.58 years (SD = 1.70). Although we
acknowledge the limitations of a student sample, we believe it is appropriate
given this is part of a new research program that is in its initial stages of testing
the model’s components.
Participants were asked which social network sites they were connected to
their partners on (they could report more than one); the most common were
Facebook (115), Snapchat (112), Instagram (81), and Twitter (43).
We asked participants a variety of questions about their relationships that
funneled them toward questions concerning situations in which their part-
ner (the observer) may have misinterpreted communication they had seen
between themselves (observed senders) and other users of the social media
site or app (observed receivers). We also asked questions about the nature of
any communication they had with their partners about what their partners
observed. The full questionnaire is available by request from the first author.
Of the 167 total respondents, 51 or 30% provided responses indicating that
hyperperception had occurred in a current or former romantic relationship
under consideration. These responses are the focus of the following analysis.
The following description of the findings is organized according to the key
components of the hyperperception model: the channel, the observed sender,
and the observed receiver. It should be noted that in this preliminary study
we did not examine the feedback loop component of the model, and we do
not conduct a robust examination of differences across social media channels.
The Channel
Participant responses revealed that the affordances highlighted in the chan-
nel component of the hyperperception model do indeed allow a partner to
observe and appraise an observed sender’s interactions with others (observed
receivers). Although our inclusion criteria required that all participants share
a social media app/site with a current or former partner, and therefore all par-
ticipants had the ability to observe and appraise each other’s interactions and
relationships, we found variation in the extent to which participants reported
having done so, and we found that some of the behaviors leading to hyper-
perception were encouraged by the norms of the social media channel. Some
participants were aware that their social media interactions might be observed
by others. As P70 noted, “Social media allows people to constantly monitor
their significant others interactions.” Many of the participants’ responses indi-
cated awareness of some of the affordances described in the channel compo-
nent of the hyperperception model. In terms of accessibility, P123 described
a partner as suspicious about a specific “facebook post,” which is a relatively
public type of communication on that social media channel. P176 said that
a partner expressed jealous feelings when the partner (observer) saw other
men post on the participant’s (observed sender’s) photos. This suggests that
the participant’s partner (observer) was jealous after accessing and appraising
others’ comments (observed receivers) on the participant’s (observed send-
er’s) relatively publicly available photos.
Participant responses also suggested that the affordance of persistence
played a role in partners’ (i.e., observers’) hyperperceptions of their social
media interactions. P29 mentioned that a partner expressed feeling insecure
after P29 (the observed sender) posted pictures that persisted on Instagram (a
social media channel on which posts persist until the user takes them down).
P66 shared that a partner felt jealous when the partner (observer) saw P66
(the observed sender) like another person’s “old tweets” on Twitter. This is
another example of how persistence may play a role in the hyperperception
process because tweets and reactions/comments on Twitter persist on that
channel unless they are removed by the people who posted them. The third
affordance, association, also came up in several participant responses. P18 said
that a partner (the observer) confronted P18 when P18 (the observed sender)
interacted with “old Class mates from high school” (the observed receivers)
on the social media channel Facebook. P18’s partner (observer) was able to
identify observed receivers because Facebook encourages users to post infor-
mation that links users to offline personas (birth name, location, when they
went to school, and the like.). P122 (the observed sender) also said that a
partner expressed jealousy when the partner (observer) saw P122 follow and
talk to “a specific individual” (the observed receiver). P160 also explained

that a partner (the observer) gets jealous when P160 (the observed sender)
interact with P160’s “Best friend who used to have romantic interest” (the
observed receiver) in P160 because it “leads him” (the partner/observer)
“to suspect that the friend could be trying to rekindle or start something
I guess.” In this case, the observer is able to identify the observed receiver
as the observed sender’s best friend thanks to Snapchat encouraging users to
post photos and videos that feature the user’s face and location, which in turn
makes users linkable to offline corporeal personas. These examples show that
the affordances of social media channels increase the likelihood of observers
appraising content that could lead them to make hyperperceptions of their
romantic partners’ social media interactions.
The Observed Sender
We asked people who were currently in a relationship to indicate on a scale
of 1 (never)-6 (always) “How often do you look at what your partner posts
or shares online?” and “How often do you think your partner looks at what
you have posted on any social network sites or social media apps?” Most
look at their partner’s online interactions and posts frequently (M = 4.44,
SD = 1.34), and think their partner looks at their interactions and posts fre-
quently as well (M = 4.31, SD = 1.37). These averages suggest that the phe-
nomenon of wanting to know about a romantic partner’s social life described
by Dunbar (2004) persists and that using social media can facilitate this infor-
mation-seeking process.
We also asked participants (i.e., the observed sender) to describe situations
where their partners (the observers) suggested that their social media inter-
actions seemed to indicate that they were flirting or interested in a romantic
relationship with another person (an observed receiver) and if so, to describe
those situations. Our participants’ responses reveal that people’s roman-
tic partners can and do develop inflated impressions of their social media
interactions, which is consistent with the hyperperception model. Some of
these accounts suggest aspects of the observer and the relationship that may
increase the likelihood of hyperperceptions.
For example, P183 (the observed sender) said a “boyfriend” (the partner,
observer) “had thought that I was interested in someone else … due to the
fact that I like his tweets.” P183 went on to say that this partner’s perception
of P183’s Twitter likes was inaccurate; P183 did not like the tweets because
of who tweeted them (the observed receiver) but because the tweets them-
selves were “funny.” In fact, P183 said the partner went so far as to assert that
liking another person’s tweets meant that P183 was interested in that other
person. In other words, doing something as innocuous as liking a tweet can
lead a romantic partner to ask serious questions about the partner and possi-
bly the relationship.
Similarly, P15 (the observed sender) said that there was a situation where
a partner (the observer) “felt like I cheated on” him because of a conver-
sation P15 (the observed sender) had with someone else (the observed
receiver) on Facebook (the social media channel). To try to rectify the situ-
ation, P15 ended that Facebook conversation and unfriended the observed
receiver. Moreover, P15 said the partner’s perception was “very inaccurate
and it caused unnecessary frustration” in their relationship. P15 went on to
say that this Facebook conversation was “harmless and no worse than what
he’d say with his friends.” P15’s experience shows that hyperperceptions of
social media interactions can not only cause discord in a relationship, but can
also have ramifications beyond that relationship (e.g., friendships).
Another participant, P17 (the observed sender) wrote that a partner (the
observer) asked about P17’s Facebook interactions with “old Class mates
from high school” (the observed receivers) indicating that P17’s partner’s
perception of these interactions was “inaccurate.” P17 suggested that the
partner may have jumped to such hyperperceptions because he suffers from
“Low self-esteem low self-confidence” and also noted that this was the part-
ner’s 3rd marriage. This suggests that aspects of the observer such as self-es-
teem and relationship history may not only increase people’s motivation to
observe and appraise their partners’ social media interactions but also make it
more likely that observers will hyperperceive those interactions.
The Observed Receiver

According to the hyperperception model, when an observer senses that a
sender is paying attention to a particular person (receiver) on a social media
channel, the observer may develop an inflated impression of the interactional
and relational intimacy between the observed sender and observed receiver.
This is consistent with our findings: participants’ partners’ inflated impres-
sions seemed to be affected by who they were interacting with on a social
media channel. For example, P75 (observed sender) said that a partner
(observer) inaccurately interpreted P75’s social media interactions with an
ex (observed receiver), which led P75’s partner (the observer) to feel jealous.
However, P75 also said that the partner had “a right to feel jealous” so P75
tried to reassure the partner that the ex “was history.” Perhaps P75’s partner
worries that P75 wants to get back with her ex, so seeing them interact on
social media might lead the partner to worry whether P75 would rather get
back together with the ex than stay in the present relationship.
According to the hyperperception model, the less well the observer knows
the receiver, the less likely the two are connected on- and offline, which in
turn may make it difficult for the observer to contextualize and accurately
interpret the sender’s social media interactions with that receiver. Some of our
participants mentioned that their partners were especially bothered by their
interactions with an unknown person on a social media channel. For example,
P37 (the observed sender) mentioned that a partner (the observer) was sus-
picious and jealous of P37’s interactions with “guys that he” (the observer/
partner) “doesn’t know or hasn’t met” (observed receivers). P37’s partner
was specifically suspicious of why P37 “commented on a family friend’s pic-
ture saying he looked good after losing a lot of weight.” In this case, a benign
comment meant to congratulate someone on meeting his health goals was
inaccurately interpreted as a flirtatious advance. If P37’s partner was unable to
access this family friend’s social media profile and interactions, he would not
easily be able to determine whether this observed receiver was in a relation-
ship (with someone besides P37) or that others were also congratulating this
person on his weight loss. In other words, P37’s partner would have a harder
time contextualizing P37’s interactions with this receiver, which may have
explained why he hyperperceived P37’s Facebook compliment.
Participants also discussed times when a social media interaction would be
perceived differently offline versus on social media, which is consistent with
the assertion that the less an observer is able to contextualize the observed
sender-receiver interactions, the more likely the observer may make a hyper-
perception. We asked participants if their “partner seems to make similar
judgments about” their “face-to-face interactions in comparison to” their
“social network site or social media app interactions.” P119 (the observed
sender) replied that the partner did not jump to the same conclusions when
appraising P119’s offline interactions, “because he” (the observer/partner)
“can read my facial expressions and watch my body language,” which indi-
cates the importance of nonverbal cues to indicate flirting versus mere friend-
liness. Another important contextualization cue mentioned by a participant
regarded the length of the relationship. When asked whether respondents’
partners (observers) had an accurate perception of how participants inter-
acted with others in online environments, P37 (observed sender) replied,
“not when we were first dating, but as years have passed yes.” This suggests
that hyperperceptions may be more likely to occur in early phases of relation-
ships when romantic partners have less information about how their partners
interact with various observed receivers online and thus are less able to con-
textualize and accurately interpret their partner’s interactions with others.
Anticipated and Unanticipated Consequences: When

Hyperperceptions Lead to Meaningful Conversations
According to our initial examination of the channel, observed sender, and
observed receiver components of the hyperperception model, we found that
about 30% of social media users seem to hyperperceive a romantic partner’s
social media interactions with other people outside the relationship. This
indicates that our survey methodology was successful. It also suggests that
the unique affordances of social media channels can lead observers to develop
hyperperceptions or inflated impressions of an observed sender’s interactional
intimacy with others outside the relationship.
When people are confronted by their romantic partners about their social
media interactions, they might assume there will be an argument, possible
harm to the relationship, perhaps even the beginning of the end of that rela-
tionship. This seems especially likely if the partner (observer) makes accusa-
tions that one’s (sender) social media interactions suggest that the observed
sender is pursuing or being pursued by a potential romantic rival. It was sus-
pected this would cause discord in the relationship—and it did. The descrip-
tion of participants’ (observed senders) responses above shows how partners
(observers) engage in confrontation communication and express feelings
such as anxiousness and jealousy when observers sense their partners are pur-
suing or being pursued by other users (observed receivers) on a social media
channel.
However, that is not necessarily the case. As noted above, the major-
ity of our participants reporting hyperperceptions (36 of 51, or 71%) indi-
cated that talking about the seemingly suspicious social media interactions
ultimately benefitted their romantic relationship. Only six reported negative
outcomes from talking about their partner’s hyperperceptions. In report-
ing a positive outcome, P15 (observed sender) said that a partner (observed
receiver) assumed P15 was cheating on him based on his observations of
P15’s Facebook interactions with an observed receiver. P15 indicated that
talking about the issue with the partner made the partner “aware that if some-
thing bothers” him he has “to say something about it right then and there.”
P15 also said that talking this through with the partner made the relationship
“stronger because of” their “open communication and appreciation of how
each person felt about the situation.”
Several other participants expressed similar sentiments. P37 said that

talking with a partner about his hyperperceptions improved the communi-
cation within their relationship, and that P37’s partner asks “questions now
instead of assuming” what is happening on P37’s social media. P75 said talking
about a partner’s hyperperceptions has helped the couple grow “closer” and
engender “more trust” in their relationship. P106 said that talking about a
partner’s past hyperperceptions helped them strategize to avoid future hyper-
perceptions: they now “have boundaries and rules” about “engaging in oppo-
site sex online interactions.” P155 explained how talking about the partner’s
hyperperception benefitted the relationship:
it has made it easier to trust him because after our talk i got to see how he felt
about the situation and it made us realize that we both would never want to
upset each other or make each other feel any way. however the conversation is
staying open because we want to be conformable and not feel bad for having
“doubts” if you can call it that about our relationship. its made it easier to talk
about other hard topics as well.
In summary, when romantic partners talk about their hyperperceptions of

each other’s social media interactions, the result can be positive: it can lead
to more open communication, more trust in the other, and perhaps even
decrease the likelihood of future hyperperceptions.
Future Directions
In addition to examining what happens in romantic relationships when one’s
partner seems to be hyperperceiving one’s own social media interactions, the
hyperperception model can be applied to other romantic and platonic phe-
nomena. In other research, we have used this model to examine what leads
observers to make hyperperceptions of their romantic partners’ social media
interactions (Spottswood & Carpenter, 2019b) and also to assess how hyper-
perceptions can have a variety of different effects on how romantic partners
perceive their own and each other’s social media interactions (Spottswood
& Carpenter, 2019a). Within the context of romantic relationships, future
research could explore what specific behaviors lead people to hyperperceive
their partner’s social media interactions. This may be especially aided by sur-
veying or interviewing both the observer and the observed sender simultane-
ously to see what specifically leads to hyperperceptions.
The hyperperception model can also be applied to other types of rela-
tionships where the overall impression of others’ social media interactions in
comparison with one’s own leads the observer to feel lonely, hurt, perhaps
even socially anxious and depressed. It can be a good tool to explain how
and why social media use can incite jealousy or hurt in potential, current, and
terminated romantic relationships, and to examine other kinds of social media
phenomena such as the negative effects of passive internet use (PIU). PIU is
the “consumption” of online content, defined as “the monitoring of all of
the content that is not specifically targeted at a given user, including friends’
broadcasts to wide audiences (e.g., status updates), or public conversations
by the user’s friend with others” (Burke, Marlow, & Lento, 2010, p. 1909–
1910). Research has shown that PIU can exacerbate feelings of loneliness,
social anxiety, and/or depression (Appel, Gerlach, & Crusius, 2016; Chen,
Fan, Liu, Zhou, & Xie, 2016; Shaw, Timpano, Tran, & Joorman, 2015). Are
these harmful psychological effects of PIU caused by people hyperperceiv-
ing their family members, friends, and other important people’s social media
interactions and relationships? The digital and cultural properties of social
media channels may lead friends, family members, and others to hyperper-
ceive each other’s interactions, which could result in them feeling left out
and lonely. The hyperperception model could be used to explore whether
hyperperceiving friends’ and other important social relations’ social media
interactions is related to feelings of loneliness, exclusion, social anxiety, or
depression.
The possible applications of the model are endless; given the ubiquity
and importance of social media, it is important that we understand why and
how hyperperceptions occur, and when discussing them can actually help
rather than hurt people’s view of themselves, those they are close to and their
relationships.
References
Anderson, T. L., & Emmers-Sommer, T. M. (2006). Predictors of relationship satisfaction
in online romantic relationships. Communication Studies, 57, 153–172.
Appel, H., Gerlach, A. L., & Crusius, J. (2016). The interplay between Facebook use,
social comparison, envy, and depression. Current Opinion in Psychology, 9, 44–49.
boyd, d. (2010). Social network sites as networked publics: Affordances, dynamics, and
implications. In Z. Papacharissi (Ed.). A networked self: Identity, community, and cul-
ture on social network sites (pp. 47–66). New York, NY: Routledge.
Brody, N. (2013). Absence—and mediated communication—makes the heart grow
fonder: Clarifying the predictors of satisfaction and commitment in long-distance
friendships. Communication Research Reports, 30, 323–332.
Burke, M., Marlow, C., & Lento, T. (2010, April). Social network activity and social
well-being. In Proceedings of the SIGCHI conference on human factors in computing
systems (pp. 1909–1912). Atlanta, GA: ACM.
Buss, D. M. (1988). From vigilance to violence: Tactics of mate retention in American

undergraduates. Ethnology and Sociobiology, 9, 291–317.
Carpenter, C. J., & Spottswood, E. L. (2019). Extending the hyperpersonal model to observ-
ing others: The hyperperception model. Manuscript submitted for publication.
Chen, W., Fan, C. Y., Liu, Q. X., Zhou, Z. K., & Xie, X. C. (2016). Passive social net-
work site use and subjective well-being: A moderated mediation model. Computers in
Human Behavior, 64, 507–514.
Dunbar, R. I. (2004). Gossip in evolutionary perspective. Review of General Psychology,
8(2), 100–110.
Evans, S. K., Pearce, K. E., Vitak, J., & Treem, J. W. (2016). Explicating affordances: A con-
ceptual framework for understanding affordances in communication research. Journal
of Computer-Mediated Communication, 22, 35–52.
Fox, J., & Vendemia, M. A. (2016). Selective self-presentation and social comparison
through photographs on social networking sites. Cyberpsychology, Behavior, and Social
Networking, 19(10), 593–600.
Gonzales, A. L., & Hancock, J. T. (2011). Mirror, mirror on my Facebook wall: Effects
of exposure to Facebook on self-esteem. Cyberpsychology, Behavior, and Social
Networking, 14, 79–83.
Hogan, B. (2010). The presentation of self in the age of social media: Distinguishing
performances and exhibitions online. Bulletin of Science, Technology & Society, 30,
377–386.
McEwan, B., & Zanolla, D. (2013). When online meets offline: A field investigation of
modality switching. Computers in Human Behavior, 29, 1565–1571.
Muise, A., Christofides, E., & Desmarais, S. (2009). More information than you ever
wanted: Does Facebook bring out the green-eyed monster of jealousy? CyberPsychology
& Behavior, 12, 441–444.
Patil, S. (2012, February). Will you be my friend?: Responses to friendship requests from
strangers. In Proceedings of the 2012 iConference (pp. 634–635). New York, NY: ACM.
Shaw, A. M., Timpano, K. R., Tran, T. B., & Joormann, J. (2015). Correlates of Facebook
usage patterns: The relationship between passive Facebook use, social anxiety symp-
toms, and brooding. Computers in Human Behavior, 48, 575–580.
Spottswood, E. L., & Carpenter, C. J. (2019a, November). Breaking up is hard to do, espe-
cially when Facebook won’t let you. Paper presented at the National Communication
Association Convention, Baltimore, MD.
Spottswood, E. L., & Carpenter, C. J. (2019b) Facebook jealousy: A hyperperception perspec-
tive. Manuscript submitted for publication.
Spottswood, E. L., & Hancock, J. T. (2016). The positivity bias and prosocial deception
on Facebook. Computers in Human Behavior, 65, 252–259.
Utz, S. (2015). The function of self-disclosure on social network sites: Not only inti-
mate, but also positive and entertaining self-disclosures increase the feeling of connec-
tion. Computers in Human Behavior, 45, 1–10.
Walther, J. B. (1996). Computer-mediated communication: Impersonal, interpersonal,

and hyperpersonal interaction. Communication Research, 23, 3–43.
Walther, J. B. (2007). Selective self-presentation in computer-mediated communica-
tion: Hyperpersonal dimensions of technology, language, and cognition. Computers
in Human Behavior, 23, 2538–2557.
12.
Proposing a Model of Social Media
Use and Well-Being
Marina Krcmar, Drew P. Cingel, Yifan Zhao, and
Lauren Taylor
For many years, Facebook was the most frequently used social media plat-
form among adolescents and young adults (Verduyn et al., 2015), although
other platforms have reached similar usage rates among 18–24-year-olds
(Lomas, 2017). However, research on the use of individual platforms such as
Instagram (Ahadzadeh, Sharif, & Ong, 2017; Kleemans, Daalmans, Carbaat,
& Anschütz, 2018), Snapchat (Utz, Muscanell, & Khalid, 2015), Twitter
(Pittman & Reich, 2016), and Facebook (Cingel & Olsen, 2018) may be
limited in its generalizability, given marked differences in affordances of vari-
ous apps. Few studies have adopted a comparative approach to understanding
differences in uses and effects across social media platforms (e.g., Utz et al.,
2015; Waterloo, Baumgartner, Peter, & Valkenburg, 2018). Concepts of
social media use, processing, and effects have yet to be thoroughly and con-
sistently explicated into one model that attempts to account for differences in
various features, use patterns, and processing differences.
Although research on social media effects has examined outcome vari-
ables as disparate as political engagement (Pennington, Winfrey, Warner,
& Kearney, 2015), adolescent development (Cingel & Krcmar, 2014), and
consumer purchase intentions (Wang, Yu, & Wei, 2012), a significant focus
has been on how social media affect aspects of individuals’ well-being. Well-
being is a broad term encompassing a number of indicators; in the context
of explicating the relationship between social media use and well-being, we
will examine studies whose dependent measures are considered dimensions of
well-being: stress/anxiety, loneliness/social connectedness, self-esteem, life
satisfaction, and body satisfaction (Bevan, Gomez, & Sparks, 2014). We will
use these well-being indicators as a set of outcome measures as a test-case of
196 Marina Krcmar, et al.
sorts, to explore which aspects of social media best help us examine effects.
Our goal is to present a preliminary model that summarizes, synthesizes, and
organizes the variables related to social media use, processing, and effects. We
are guided by four main areas of literature.
First, we examine research into how demographics such as age and sex
are related to social media use patterns. The uses and gratifications framework
has laid out the importance of motivations for media use in terms of uses and,
ultimately, effects, but has yet to be tied into an overall model of social media
use, processing, and effects.
A second area of research suggests that the specific features or affordances
of social media platforms matter in terms of effects, and more precisely, that
there are meaningful differences in outcomes for exposure to images and
image manipulations in comparison to text. We will use this literature to high-
light the importance of considering different features when studying effects
of social media use.
A third research area examines how passive versus active use of social
media differ in terms of outcomes, although we note that overall the results
are inconsistent. We believe this may be a function of how use is operational-
ized. In our discussion of active and passive use, we propose a clarification of
active and passive social media use that considers both behavioral and cogni-
tive social media activity. To our knowledge, this has not yet been done and
represents a key contribution of our model.
Finally, although not much research has examined the nature of who indi-
viduals connect or interact with on social media, there is some evidence that
communicating with close others can have more beneficial outcomes than
communicating with unknown others. Thus, what can broadly be termed
as relational proximity, or the nature or closeness of the social network, is
included as a moderator in the proposed model.
Our goal is to combine these areas of research on social media and
well-being and propose a model of social media use, processing, and effects.
Although Krause, Baum, Baumann, and Krasnova (2019) proposed a model
on social media and self-esteem, it does not broadly examine uses, processes,
and effects. In our proposed model, we argue that individual differences
(e.g., demographics and personological variables) predict the motivations for
seeking out social media, and specifically whether image- or text-based social
media are used. We argue that active social media behaviors (e.g., posting) and
active social media cognitive processing strategies (e.g., social comparison, self
presentation) yield different outcomes both from each other (e.g., behavior
vs. cognition) and from passive social media use behaviors (e.g., browsing)
as well as from passive cognitive processing (e.g., vegging out). The model
Proposing a Model of Social Media Use and Well-Being 197
presents social proximity between users as a moderator, and considers several

well-being indicators as outcomes.
Uses and Gratifications: Predicting Social Media Use

One important starting place when considering the effects of social media is the
mere fact that different apps, features, and functions within apps appear to be
used by different groups of people, perhaps with different goals regarding use.
Unlike other media communication paradigms that emphasize either media
content or media effects, uses and gratifications research focuses on media use,
asking what motivates various kinds of media use. Media users are understood
to be active, purposeful, and selective in their media choices. The classic uses
and gratifications precis states that media use is motivated by: “the social and
psychological origins of needs which generate expectations of the mass media
and other sources which lead to differential patterns of media exposure (or
engagement in other activities) resulting in needs gratifications and other con-
sequences, perhaps mostly unintended ones” (Katz, Blumler, & Gurevitch,
1973, p. 20). Later refinements of the framework differentiated between user
gratifications that are sought and those that are obtained, noting that the latter
do not necessarily correspond with the former and unmet needs may in fact
have negative and unintended effects (Palmgreen, Wenner, & Rayburn, 1980).
By understanding why we choose the media we do and how we use it,
researchers can better comprehend the entire media use process from selec-
tive exposure to media outcomes and effects. Clearly, this is important when
trying to understand the relationship between social media use and well-be-
ing. Uses and gratifications theorizing, coupled with empirical evidence, sug-
gests that motivations for use can influence effects, but then effects (such as
on self-esteem) may influence how we process further incoming information
and what we choose to consume. In other words, users’ motivations to use a
particular social media platform can influence the entire social media use-pro-
cessing-effects model both transactionally and recursively.
Several studies examine the uses and gratifications of different social
media platforms. For example, Pai and Arnott (2013) identified belonging,
hedonism, self-esteem, and reciprocity as key motivators of Facebook use.
Quan-Haase and Young (2010) found that Facebook users are motivated by
enjoyment and knowledge about social activities within their social networks,
and cited six key dimensions of Facebook use-motivations: pastime, affec-
tion, fashion, share problems, sociability, and social information. Therefore,
individuals are motivated to use different social media platforms for different
reasons (e.g., Raacke & Bonds-Raacke, 2008).
However, generalizations about the motivations of social media use are

limited because most research does not typically compare use motives across
various apps, and differences in app use patterns suggest that motives may vary
depending on both individual differences and app features (Rathnayake, &
Winter, 2018). Most broadly, it appears that images and text are used differ-
ently. For example, Malik, Dhir, and Nieminen (2016) found that users share
images on social networks to seek gratifications of affection, attention-seek-
ing, disclosure, habitual pastime, information-sharing, and social influence.
This use of image-sharing is likely because images can provide greater imme-
diacy and intimacy than text-based communication can.
In addition to being used differently, social media platforms attract differ-
ent users. For example, Perrin and Anderson (2019) discerned age-related pat-
terns: Snapchat and Instagram are used by 62% and 67% of young adults (aged
18–29), respectively, but Facebook is used by people representing more broad
age groups. Perrin and Anderson also reported other demographic differences
on social media platforms and messaging applications. For example, women
were almost three times more likely than men to use Pinterest (42% vs. 15%).
Research has shown that, like social media use in general, image-focused
behaviors are more common in females and young populations (Nesi &
Prinstein, 2015; Rosen, Stefanone & Lackaff, 2010; Stefanone & Lackaff,
2009). Further, close friends and family members are also likely to use imagery
when communicating with one another (Stefanone & Lackaff, 2009). People
from individualistic cultures are more likely to use visual communication than
are people from collectivist cultures (Rosen et al., 2010). Offline social sup-
port networks also factor into image use; individuals with a greater number
of close friends and family members share more images online than do those
with fewer friends and connections (Stefanone, Lackaff, & Rosen, 2011).
This could be because individuals are most likely to use graphics in commu-
nication when motivated by relationship maintenance concerns (Hunt, Lin,
& Atkin, 2014), and people with strong networks tend to desire relationship
maintenance over other relational concerns.
When combined, these findings clearly suggest differing motivations for
social media use stemming from demographic differences among users and
platform features, and influencing not only the platforms used by the individ-
ual, but also the communicative activities that occur via the platform.
Social Media Effects: Images vs. Text

As noted, individual differences and differences in motivations sought can
influence use of images or text on social media platforms. Thus, investigating
image effects and text effects can help us understand the link between spe-
cific aspects of social media use and well-being. Text-based communication
consists of behaviors such as sharing status updates, sending direct messages,
or commenting on others’ text or image posts; image-based communication
(visual communication) is primarily the sharing of images or videos (Highfield
& Leaver, 2016).
Although most social media platforms support sharing images and text,
broadly speaking, platforms can be categorized as primarily text- or image-
based. On text-based platforms such as Facebook and Twitter, content is
created predominantly through sharing written communication. Although
offering the option to post images, the default format is a written text post.
In contrast, the only option to create content on image-based platforms such
as Instagram and Snapchat is to share an image. In general, it seems likely that
the use-motivations and effects differ based on text vs. images, regardless of
platform. Thus far, little research has explicitly compared the uses and effects
of images and text, although research has implicitly supported such compar-
isons by virtue of the platforms under consideration. Overall, the findings
suggest that motivations for use may influence whether text or images are
used, and whether images are edited.
In terms of images and editing of these images, young users rely more
heavily on graphics than text to communicate (Grieve, 2017), tend to check
image-based platforms more frequently than they check text-based platforms,
and engage in image editing (Boyle, Earle, LaBrie, & Ballou, 2017). Further,
roughly three-quarters of all photographs shared on Instagram are filtered or
retouched (Hochman & Manovich, 2013).
Research studying well-being outcomes from the use of image-based
communication suggest that visual communication can offer a more direct,
intimate, and immediate connection to others, better enabling synchronic-
ity and a sense of shared confiding than is afforded by text communication
(Pittman & Reich, 2016). Specifically, Pittman and Reich (2016) found that,
when taking advantage of the intimate and direct nature of image-based com-
munication, young adults experienced decreased feelings of depression and
loneliness and increased happiness and life satisfaction because of a sense of
closeness to those with whom they are communicating. Similarly, users expe-
rience positive well-being outcomes when communicating visually, especially
with known individuals such as family members and friends (Lup, Trub, &
Rosenthal, 2015). Thus, a small body of work has found that image-based
social media are used to satisfy needs for social and emotional connection,
and that can encourage positive outcomes. This need (gratification sought)
and resulting need-satisfaction (gratification obtained) is likely to strengthen
positive well-being outcomes as much as the image-based content itself.

Clearly, then, motives for use interact with content to influence effects.
Although positive outcomes are certainly possible, image-focused social
media platforms can also evoke more body image concerns and related neg-
ative outcomes than do text-based platforms. This outcome is attributed to
commonly used photo-filtering and retouching practices, as well as content
selection practices emphasizing positive, polished situations that often trigger
upward social comparisons (Lup et al., 2015). Kleemans et al. (2018) sug-
gested that exposure to filtered and retouched images on social media can
lower body satisfaction among adolescents. Image-focused internet use in
general has been linked to decreased body satisfaction and increased inter-
nalization of the thin-ideal (Bair, Kelly, Serdar, & Mazzeo, 2012). Meier and
Gray (2014) found that increased exposure to images on Facebook—not
overall usage of Facebook as a whole—led to increases in weight dissatisfac-
tion, drive for thinness, internalization of the thin-ideal, and self-objectifica-
tion among adolescent girls. Thus, visual communication can affect several
facets of body-related concerns regardless of platform.
In general, Instagram usage has been linked to increases in depressive
symptoms (Lin et al., 2016; Nesi & Prinstein, 2015), as has following a large
number of strangers on Instagram. This finding was particularly strong for
those who have higher social comparison tendencies (Lup et al., 2015), espe-
cially when users are exposed to positively valenced images posted by strang-
ers (de Vries, Möller, Wieringa, Eigenraam, & Hamelink, 2018). In addition,
Utz et al. (2015) found that Snapchat use elicited higher levels of jealousy
between love interests than did Facebook use. Although the researchers did
not look at images versus text specifically, it is possible that stand-alone images
without much text, as occurs on Snapchat more than on Facebook, may have
negative consequences, including upward social comparison, similar to the
negative effects found for Instagram.
Overall, these findings suggest that users are motivated by different rea-
sons to use or interact with images as compared to text, and that effects on
well-being are contingent on the way in which users interact with the image-
or text-based content. In the following section, we review research pertaining
to active or passive use of social media, and identify some key mechanisms
influencing well-being effects. Specifically, we identify active and passive use
as occurring both behaviorally and cognitively.
Active and Passive Use

In both communication and psychology research, the earliest instruments
measuring social media focused on extent of use rather than on how social
media were used. The research rarely accounted for the variety of activities
afforded by different platforms and therefore did not encourage cross-plat-
form comparisons. Olufadi (2016) highlighted the need for better measure-
ment that addressed both the motivations for and uses of social media, and
the affordances of different platforms. In order to better address this issue, we
present a clear definition of social media use by reviewing literature on active
and passive use of media, refining and specifying the existing definitions, and
linking research on social comparison theory and self-presentation to active
cognitive use.
Early in the current decade, researchers began to dichotomize social media
use into passive use and active use (Burke, Marlow, & Lento, 2010), with
special focus on social media features. For example, Krasnova, Wenninger,
Widjaja, and Buxmann (2013) proposed a behavioral definition of active use
behaviors as communication promoting the exchange of information, such
as posting status updates and sending direct messages; passive use behaviors
imply browsing behaviors, such as viewing others’ posts and liking others’
photos. Another approach is to explore cognitively active social media use,
where social comparison and self presentation have consistently emerged as
important variables (e.g., Kleemans et al., 2018).
We acknowledge that social comparison to social media images, posts,
and information has not been included in traditional definitions of active use,
but argue here that cognitive social comparison should be considered a kind
of mental activity that differs from behaviorally active use, and that these use
differences often result in important and different effects. Thus, when consid-
ering active and passive use, we must distinguish both between active and pas-
sive behaviors and more or less active cognitive processing. Although the active
and passive mechanisms identified in the literature differ across studies, here
we identify behaviorally active posting, interaction (defined by Yang (2016)
as direct communication with other users’ involvement), and broadcasting
(defined by Yang (2016) as disseminating information to a general audience
without mentioning specific users) as active behavior.
Also, we define cognitive social comparison and self-presentation as active
cognitive processing. In contrast, behavioral passivity occurs through brows-
ing (defined by Yang (2016) as viewing other people’s profiles or homepages
without making any comments), and cognitive passivity is associated with
vegging out.
Social Comparison and Self-Presentation

Social comparison theory (Festinger, 1954) argues that people tend to
evaluate their opinions and abilities against others’, and that downward
comparison—when we compare ourselves to others we perceive to be doing

worse than ourselves—results in increased self-esteem and perceived well-be-
ing. In contrast, upward comparison—to an individual perceived to be doing
better than us—results in decreased self-esteem and perceived well-being.
Both forms of comparison require mental activity: processing the information
provided by another person and comparing oneself against that information.
As such, we identify social comparison as cognitively active social media use,
and we review the relevant literature in light of this.
For example, Steers, Wickham and Accitelli (2014) found that time spent
on Facebook was positively correlated to social comparison. Furthermore,
upward social comparison to what they called social media “highlight reels,”
reflecting people’s tendency to share “only positive and/or self-enhancing
news but not fully disclose their daily struggles” (p. 725), resulted in increased
depressive symptoms among users. Thus, active social comparison seems to
have negative outcomes. Similarly, even when social comparison to social
media is not explicitly measured, negative effects have been found. For exam-
ple, viewing others’ social media has been associated with diminished self-es-
teem (Frison & Eggermont, 2016; Mehdizadeh, 2010; Wilson, Fornasier,
& White, 2010). In contrast, viewing one’s own profile seems to increase
self-esteem (Gonzales & Hancock, 2011), perhaps by allowing for reflection
on one’s own positive moments rather than engaging in social comparison.
Self-presentation is another concept that has been associated with social
media use and outcomes. Self-presentation is simply engaging in behaviors
(e.g., reflecting on and posting text and images) that serve to shape one’s
online identity. Specifically, selective self-presentation appears to encourage
an identity shift such that the users’ self-perception becomes more consistent
with what has been presented online (Carr & Hayes, 2019). Therefore, selec-
tive self-presentation may serve to enhance self-esteem given that self-pre-
sentation is typically strategically positive. This may be related to Pounders,
Kawalcyzk, and Stowers’ (2016) finding that selective self-presentation is typi-
cally associated with individuals having low self-esteem to begin with. Overall,
therefore, it appears that cognitively active behaviors such as social compar-
ison and self-presentation are generally associated with negative well-being
outcomes, with some exceptions such as when identity matching to positive
self-presentation occurs.
Social Media and Relational Proximity

Although the body of research is small, it suggests that the relational proximity
(relative emotional closeness or distance) from the individual to the perceived
audience moderates the relationship between processing variables (various

types of active and passive use) and well-being outcomes. Close friends and
family members use images more when communicating with one another
(Stefanone & Lackaff, 2009) than strangers do, and these images seem to
increase perceived relational closeness. Images are also likely to be used by
people with large and strong offline social support networks (Stefanone et al.,
2011), and users are most likely to use visuals and graphics in communication
when motivated by relationship maintenance concerns (Hunt, Lin & Atkin,
2014). Finally, Lup et al. (2015) found that social media use is related to
well-being for those who communicate primarily with close others as com-
pared to with strangers. Given the interplay between images, relational close-
ness and positive well-being outcomes (Pittman & Reich, 2016), we argue
that the relationship between the user and the target of communication is an
important—and likely moderating—variable in predicting outcomes.
Direction of Causality
In addition to the content areas discussed thus far, our model also attends
to the nature of the relationships among these content areas. Most studies
on social media are cross-sectional and, although valuable, do not illuminate
the direction of influence. For instance, Yang (2016) found that the active
behavior of direct communication with other users, as well as the passive
behavior of browsing, were negatively related with loneliness. On the other
hand, active broadcasting behavior was positively related to loneliness. This
finding is interesting given its slight departure from the general conclusion
that passive use has problematic outcomes and active use has positive out-
comes. We argue that these findings suggest that in some cases, such as when
social media are used for broadcasting, use may be predicted by loneliness; on
the other hand, interacting with others may actually result in the alleviation
of loneliness.
Other well-being research, in this case on the relationship between stress/
anxiety and social media use, has also yielded mixed results, reinforcing the
need to address the direction of causality. Although Bevan et al. (2014) found
a positive relationship between time spent on Facebook and stress, Shaw,
Timpano, Tran, and Joormann (2015) reported that social anxiety symptoms
were only associated with passive Facebook use. As with the findings con-
cerning loneliness, the direction of the relationship between social media use
(either active or passive use) and anxiety and stress might function differently
depending on the social media activity and the outcome. That is, stress or
loneliness may prompt social media use in some cases, but may be the effect of
social media use in others. Consider, for example, a social media user who is
more anxious or stressed than other people. This user might find face-to-face
interaction uncomfortable and challenging, and thus prefer online commu-
nication. However, replacing face-to-face communication with social media
may lead to a more detrimental result; that is, for socially anxious individuals,
spending more time on social media may exacerbate social anxiety (Weidman
et al., 2012).
Therefore, well-being can serve as either a cause or an effect of social
media use, and may well be transactional or recursive. For example, some-
times loneliness may encourage social media use and then be alleviated by it,
such as in cases of active behavioral communication; however, in other cases,
such as when lonely users passively engage in browsing behavior, social media
use may exacerbate loneliness. In this case, the relationship may be recursive.
Thus, any model of social media use, processing, and effects must consider
type of behavioral and cognitive processes and also examine potential transac-
tional and recursive possible relationships.
Proposing an Initial Model of Social Media Use and Well-Being

The emerging evidence discussed above shows mixed and at times conflicting
results when scholars investigate the relationship between social media use
and variables of well-being without considering the many interacting factors
associated with the user and a platform’s affordances. A careful analysis of dif-
ferent patterns of relationships suggests that several variables are crucial: types
of use (active and passive use; both behavioral and cognitive); image-based
vs. text-based features; the relational distance between the user and target of
social media; and directions of causality.
We need to examine these relationships using experiments and longitu-
dinal designs to untangle the causal relationships between social media use
and well-being. In the meantime, however, based on the extant literature, we
can (1) identify key variables related to use, processing, and effects; (2) orga-
nize these variables into a proposed process model; and (3) identify when
relationships are certainly unidirectional (e.g., age to social media choice),
and when they are likely to be bidirectional, transactional, or even recursive
(e.g., exposure, social comparison, and self-esteem). In developing our model
(Figure 12.1), we are aided not only by the literature reviewed in this chapter,
but also by long-supported theory such as uses and gratifications (Katz et al.,
1973) and social comparison (Festinger, 1954), both of which have been
applied to legacy and social media.
Figure 12.1: General Model of Social Media Uses, Processes, Effects and Well-Being
What generalizations, then, can be made and how can they best be inte-
grated into a model? Several findings appear to be consistent across the liter-
ature, and from those consistent findings, albeit across different studies that
examined different social media, we can make some tentative generalizations.
First, uses and gratifications can be used to frame what is known about pat-
terns of social media use in terms of demographic predictors, gratifications
sought, and behavioral activities such as posting, browsing, or passively view-
ing. The demographics of age and sex predict use patterns, perhaps because
these groups are more likely to seek out gratifications that emerge from com-
municating (e.g., social connection) and entertainment. Thus, we could argue
that a uses and gratifications perspective suggests that demographics influence
gratifications sought (e.g., social connections), which in turn prompt the use
of certain features within platforms and specific behavioral activity that is
more (e.g., communicating) or less (e.g., browsing) active.
The gratifications sought and the media selected could be grouped by
platform choices (e.g., Facebook or. Twitter); however, differences in text
exposure effects and image exposure effects suggest that image vs. text may be
a more fruitful grouping variable. Specifically, image effects appear to depend
at least in part on what is done with the image (e.g., filtering or editing) and
how the images are used (e.g., to communicate with family and friends).
Effects of text-based social media are also based on demographics, although,

again, various factors interact to influence outcomes.
Cognitive processing variables also matter. Social comparison theory
offers an important grounding not only in considering effects but also in
understanding the very concept of activity. Whereas media use behaviors (e.g.,
posting a photo) can be active or passive, and guided by uses and gratifica-
tions, processing of media content can also be conceived along a continuum
from cognitively active (e.g., social comparison) to cognitively passive (e.g.,
vegging out).
Ultimately, refining and assessing a complete model of social media uses,
processing, and effects require much additional research. What we propose
here is a first step. The model considers demographics, gratifications sought,
behavioral social media choices (e.g., posting vs. browsing), platform affor-
dances (e.g., text vs. images); relational proximity of other users; cognitive
processing variables; and ultimately well-being outcomes. Future research can
be guided by this model with an eye toward exploring the specific variables
identified in ways that are more systematic than in past research. Theoretically,
we might explore the precise mechanisms that encourage problematic social
comparison to images. Practically, we might look at how platform affordances
might encourage or discourage problematic use.
In sum, additional research may use this model as a guide but may make
revisions to refine it. In the end, any subsequent empirically supported revi-
sions to our model will allow for a better understanding of the effects of
media on well-being and for more constructive use of social media. Given
that social media use is near-universal among some populations, studying the
predictors of use, the effects of exposure, and measuring how social media are
used in terms of cognitions, behaviors, and platform affordances will become
increasingly important.
References
Ahadzadeh, A. S., Sharif, S. P., & Ong, F. S. (2017). Self-schema and self-discrepancy medi-
ate the influence of Instagram usage on body satisfaction among youth. Computers in
Human Behavior, 68, 8–16.
Bair, C. E., Kelly, N. R., Serdar, K. L., & Mazzeo, S. E. (2012). Does the Internet func-
tion like magazines? An exploration of image-focused media, eating pathology, and
body dissatisfaction. Eating Behaviors, 13, 398–401.
Bevan, J. L., Gomez, R., & Sparks, L. (2014). Disclosures about important life events
on Facebook: Relationships with stress and quality of life. Computers in Human
Behavior, 39, 246–253.
Boyle, S. C., Earle, A. M., LaBrie, J. W., & Ballou, K. (2017). Facebook dethroned: Revealing
the more likely social media destinations for college students’ depictions of underage
drinking. Addictive Behaviors, 65, 63–67.
Burke, M., Marlow, C., & Lento, T. (2010). Social network activity and social well-be-
ing. Proceedings of the SIGCHI Conference on Human Factors in Computing Systems,
1909–1912.
Carr, C. T., & Hayes, R. A. (2019). Identity shift effects of self-presentation and confirma-
tory and disconfirmatory feedback on self-perceptions of brand identification. Media
Psychology, 2(3), 418–444.
Cingel, D. P., & Krcmar, M. (2014). Understanding the experience of Imaginary Audience
in a social media environment. Journal of Media Psychology, 26, 155–160.
Cingel, D. P., & Olsen, M. K. (2018). Getting over the hump: Examining curvilinear rela-
tionships between adolescent self-esteem and Facebook use. Journal of Broadcasting
and Electronic Media, 62, 215–231.
de Vries, D. A., Möller, A. M., Wieringa, M. S., Eigenraam, A. W., & Hamelink, K.
(2018). Social comparison as the thief of joy: Emotional consequences of viewing
strangers’ Instagram posts. Media Psychology, 21, 222–245.
Festinger, L. (1954). A theory of social comparison processes. Human Relations, 7(2),
117–140.
Frison, E., & Eggermont, S. (2016). Exploring the relationships between different types
of Facebook use, perceived online social support, and adolescents’ depressed mood.
Social Science Computer Review, 34, 153–171.
Gonzales, A. L., & Hancock, J. T. (2011). Mirror, mirror on my Facebook wall: Effects
of exposure to Facebook on self-esteem. Cyberpsychology, Behavior, and Social
Networking, 14, 79–83.
Grieve, R. (2017). Unpacking the characteristics of Snapchat users: A preliminary investiga-
tion and an agenda for future research. Computers in Human Behavior, 74, 130–138.
Highfield, T., & Leaver, T. (2016). Instagrammatics and digital methods: Studying visual
social media, from selfies and GIFs to memes and emoji. Communication Research
and Practice, 2, 47–62.
Hochman, N., & Manovich, L. (2013). Zooming into an Instagram City: Reading the
local through social media. First Monday, 18.
Hunt, D. S., Lin, C. A., & Atkin, D. J. (2014). Communicating social relationships via the
use of photo-messaging. Journal of Broadcasting and Electronic Media, 58, 234–252.
Katz, E., Blumler, J. G., & Gurevitch, M. (1973). Uses and gratifications research. Public
Opinion Quarterly, 37, 509–523.
Kleemans, M., Daalmans, S., Carbaat, I., & Anschütz, D. (2018). Picture perfect: The
direct effect of manipulated Instagram photos on body image in adolescent girls.
Media Psychology, 21, 93–110.
Krasnova, H., Wenninger, H., Widjaja, T., & Buxmann, P. (2013). Envy on Facebook: A
hidden threat to users’ life satisfaction? Proceedings of the 11th International Conference
on Wirtschaftsinformatik, Heidelberg, Germany (pp. 230–239).
Krause, H. V., Baum, K., Baumann, A., & Krasnova, H. (2019). Unifying the detrimental
and beneficial effects of social network site use on self-esteem: A systematic literature
review. Media Psychology, 1–38. Advance online publication. https://doi.org/10.10
80/15213269.2019.1656646.
Lin, L., Sidani, J. E., Shensa, A., Radovic, A., Miller, E., Colditz, J. B., … & Primack, B.
A. (2016). Association between social media use and depression among U. S. college
adults. Depression & Anxiety, 33, 323–331.
Lomas, N. (2017). Teens favoring Snapchat and Instagram over Facebook, says eMarketer.
Retrieved from https://techcrunch.com/2017/08/22/teens-favoring-snapchat-
and-instagram-over-facebook-says-emarketer/
Lup, K., Trub, L., & Rosenthal, L. (2015). Instagram #Instasad?: Exploring associations
among Instagram use, depressive symptoms, negative social comparison, and strang-
ers followed. Cyberpsychology, Behavior, and Social Networking, 18, 247–252.
Malik, A., Dhir, A., & Nieminen, M. (2016). Uses and gratifications of digital photo shar-
ing on Facebook. Telematics and Informatics, 33, 129–138.
Mehdizadeh, S. (2010). Self-Presentation 2.0: Narcissism and self-esteem on Facebook.
Cyberpsychology, Behavior, and Social Networking, 13, 357–364.
Meier, E. P., & Gray, J. (2014). Facebook photo activity associated with body image
disturbance in adolescent girls. Cyberpsychology, Behavior, and Social Networking, 17,
199–206.
Nesi, J., & Prinstein, M. J. (2015). Using social media for social comparison and feedback
seeking: Gender and popularity moderate associations with depressive symptoms.
Journal of Abnormal Child Psychology, 43, 1427–1438.
Olufadi, Y. (2016). Social networking time use scale (SONTUS): A new instrument for
measuring the time spent on the social networking sites. Telematics and Informatics,
33, 452–471.
Pai, P., & Arnott, D. C. (2013). User adoption of social networking sites: Eliciting uses
and gratifications through a means–end approach. Computers in Human Behavior,
29(3), 1039–1053.
Palmgreen, P., Wenner, L. A., & Rayburn, J. D. (1980). Relations between gratifications
sought and obtained: A study of television news. Communication Research, 7(2),
161–192.
Pennington, N., Winfrey, K. L., Warner, B. R., & Kearney, M. W. (2015). Liking Obama
and Romney (on Facebook): An experimental evaluation of political engagement
and efficacy during the 2012 general election. Computers in Human Behavior, 44,
279–283.
Perrin, A., & Anderson, M. (2019). Share of US adults using social media, including
Facebook, is mostly unchanged since 2018. Pew Research Center. Retrieved from
https://www.pewresearch. org/fact-tank/2019/04/10/share-of-us-adults-using-
socialmedia-including-facebook-is-mostly-unchanged-since-2018
Pittman, M., & Reich, B. (2016). Social media and loneliness: Why an Instagram picture
may be worth more than a thousand Twitter words. Computers in Human Behavior,
62, 155–167.
Pounders, K., Kowalczyk, C. M., & Stowers, K. (2016). Insight into the motivation
of selfie postings: Impression management and self-esteem. European Journal of
Marketing, 50(9/10), 1879–1892.
Quan-Haase, A., & Young, A. L. (2010). Uses and gratifications of social media: A com-
parison of Facebook and instant messaging. Bulletin of Science, Technology & Society,
30(5), 350–361.
Raacke, J., & Bonds-Raacke, J. (2008). MySpace and Facebook: Applying the uses and
gratifications theory to exploring friend-networking sites. Cyberpsychology, Behavior,
and Social Networking, 11, 169–174.
Rathnayake, C., & Winter, J. S. (2018). Carrying forward the uses and grats 2.0
agenda: An affordance-driven measure of social media uses and gratifications. Journal
of Broadcasting & Electronic Media, 62(3), 371–389.
Rosen, D., Stefanone, M. A., & Lackaff, D. (2010). Online and offline social net-
works: Investigating culturally-specific behavior and satisfaction. Proceedings of the
Hawaii International Conference on System Sciences, Kaloa, Kauai, Hawaii.
Shaw, A. M., Timpano, K. R., Tran, T. B., & Joormann, J. (2015). Correlates of Facebook
usage patterns: The relationship between passive Facebook use, social anxiety symp-
toms, and brooding. Computers in Human Behavior, 48, 575–580.
Steers, M. N., Wickham, R. E., & Acitelli, L. K. (2014). Seeing everyone else’s highlight
reels: How Facebook usage is linked to depressive symptoms. Journal of Social and
Clinical Psychology, 33(8), 701–731.
Stefanone, M. A., & Lackaff, D. (2009). Reality television as a model for online
behavior: Blogging, photo, and video sharing. Journal of Computer-Mediated
Communication, 14, 964–987.
Stefanone, M. A., Lackaff, D., & Rosen, D. (2011). Contingencies of self-worth and
social networking-site behavior. Cyberpsychology, Behavior, and Social Networking,
14, 41–49.
Utz, S., Muscanell, N., & Khalid, C. (2015). Snapchat elicits more jealousy than
Facebook: A comparison of Snapchat and Facebook use. Cyberpsychology, Behavior,
and Social Networking, 18, 141–146.
Verduyn, P., Lee, D. S., Park, J., Shablack, H., Orvell, A., Bayer, J., … Kross, E. (2015).
Passive Facebook usage undermines affective well-being: Experimental and longi-
tudinal evidence. Journal of Experimental Psychology, 144, 480–488. doi: 10.1037/
xge0000057
Wang, X., Yu, C., & Wei, Y. (2012). Social media peer communication and impacts on
purchase intentions: A consumer socialization framework. Journal of Interactive
Marketing, 26(4), 198–208.
Waterloo, S. F., Baumgartner, S. E., Peter, J., & Valkenburg, P. M. (2018). Norms of online
expressions of emotion: Comparing Facebook, Twitter, Instagram, and WhatsApp.
New Media and Society, 20, 1813–1831. doi: 10.1177/1461444817707349
Weidman, A. C., Fernandez, K. C., Levinson, C. A., Augustine, A. A., Larson, R. J., &
Rodebaugh, T. L. (2012). Compensatory internet use among individuals higher in
social anxiety and its implications for well-being. Journal of Personality and Individual
Differences, 53, 191–196.
Wilson, K., Fornasier, S., & White, K. M. (2010). Psychological predictors of young adults’
use of social networking sites. Cyberpsychology, Behavior, and Social Networking, 13(2),
173–177.
Yang, C. (2016). Instagram use, loneliness, and social comparison orientation: Interact
and browse on social media, but don’t compare. Cyberpsychology, Behavior and Social
Networking, 19, 703–708.
13.
Audiences Assemble: Becoming an
Audience and Produser in Mixed
Media Environments
Annette Hill
How can we understand the patterns and contours of audiences in a mixed

media environment? I’m curious about the tracks people make as they become
aware of themselves as constituting an audience. These tracks include the way
we listen to a soundtrack and relate this to the narrative in a drama; listening
can help us find our way through storytelling, signposting how we relate to
broader themes and cultural contexts. Another set of tracks might move us
from the identity of audience to produser, for example making a remix from
a soundtrack as a creative response to broader themes in a drama, another
means of articulating our identities in the media landscape.
What follows is an argument about audience assemblage that resists
the common ways in which we conceptualize audiences as already there, an
aggregate to be measured. This argument also resists the conceptualization of
audiences as produsers (Bruns, 2008), where there is an end goal, an engage-
ment result as it were. For Tim Ingold (2011) the meaning of the term “pro-
duction” is ripe for re-assessment, shifting its meaning from the activities
of making and building to processes of hope, growing, and dwelling. Here,
we see a way of understanding audience practices as a process of becoming,
not an object or construction but what we might think of as path trails to be
found within our encounters with media.
This chapter offers a critical analysis of the creative practices of sound
production and audience experiences for the cult conspiracy television drama
Utopia, to reveal people’s awareness of themselves as becoming audiences and
produsers. What is significant for this analysis is how the soundtrack mixes
with the visual storytelling, fostering a distinctive way of listening and seeing
212 Annette Hill
not only the world depicted in Utopia but also the real world in which audi-
ences, listeners, and producers live. This audio-visual composition generates
audience assemblage, where people make connections, becoming producers
of hope, as well as produsers of new content. In the case of Utopia, these
audience assemblages, occurring in various geographical places, destablize the
market terrain for digital television, generating a geo-political audience alert
to the power relations within media industries and in society and culture.
Affect, Assemblage, and Audiences

We start with the letter A—theories and concepts relating to affect, assem-
blage, and audiences. These are all influential constructs in researching the
subjectivities of people and their identities as audiences, fans, consumers, and
produsers. Cultural and social theory on affect and emotions has tended to
generate meta level analysis and philosophical reflections on psychological
and bodily experiences, and human and non-human relations. To apply these
philosophical and, at times, esoteric writings to social scientific audience stud-
ies, which rely extensively on empirical research, poses a challenge in that
these concepts have to be worked in a more pragmatic fashion within audi-
ence research. These are the A roads of affect and assemblage that allow us to
follow audience orientations in the media landscape.
Developments in philosophy and socio-cultural theory on affect draw
attention to the highly subjective and visceral elements of human experience.
These theories want to “introduce the energetic, the physical and sensual”
(Wetherell, 2012, p. 10) back into the social sciences and humanities. For
example, Nigel Thrift (1996) used non-representational theory to prioritize
bodily experience. This non-representational approach to cultural geography
highlights the significance of practice, using theories of embodiment and affect
to explore practical knowledge and investigate what humans and non-humans
do and don’t do in physical environments. For Thrift, it is important to rein-
troduce phenomenology and the sensuous elements of cultural life into stud-
ies within geography. As we shall see, such attention to theories of practice
is significant to this analysis of audience assemblage, in particular the ways
in which people constitute themselves as audiences through their cultural
experiences.
Non-representational theory has also been influential in anthropological
research related to subjectivities. Theories of movement, rather than practice
theory, inspire Ingold’s thinking on wayfaring; he argued that research into
the production of human experience involves bringing together materials,
bodies, and flows of human experience: “coupling action and perception along
Audiences Assemble 213
paths of movement” (2011, p. 15). Ingold urged researchers to follow what is

going on, “tracing the multiple trails of becoming” (2011, p. 14). As Moores
argued (2018), the significance of Ingold’s research on a non-representa-
tional approach to media studies is to emphasize the quotidian movements of
audiences and users as they make meanings within lines of storytelling, thus
linking embodiment and affect to audience engagement with narrative and
storytelling in media landscapes.
Other aspects of affect and bodily experiences relate to emotion and
labor. In sociology, the early work of Arlie Hochschild (1979) on caring pro-
fessions proved influential in understanding emotional labor; she studied bill
collectors and airline flight attendants, who had a nuanced sense of the social
context to what she called emotion work and the management of emotional
labor, for example keeping a calm demeanor on a transatlantic flight. When
these feelings rules are broken, order must be restored before the emotional
environment is restabilized. Management of emotions is structured by social
norms and values, in particular in relation to gender, class, and race. In her
later work, Hochschild (2003) connected what she called the managed heart
with the neo-liberalization of intimate life. In the sociology of emotions, the
body, affect, and emotion are transformed into an affective economy, for
example, the way women’s care work is undervalued in the commercialization
of the private sphere.
Pertinent to my analysis of audience assemblage are the connections
across affect and theories of assemblage. Manuel Delanda (2006) drew on
Deleuze and Guattari’s (1988) philosophical writings on machine-human
assemblage, where forms connect, break, and re-form into alternative flows
of action; he noted how assemblage relates to parts that fit together, not in a
uniform way, but by energetically establishing new relations between objects,
persons, structures, and environments. Delanda’s development of theories of
assemblage signals how the human and non-human, the organic and machine-
made components, connect and disconnect, sparking new and at times radical
assemblages. Thus media infrastructures such as social media platforms and
software engineering such as algorithms form institutional assemblages with
media devices and the people using these devices, which in turn generate
flows of action, breakages, and alternative assemblages of machine-human
productions. One audience-oriented example of this is seen in the bio-labor
of micro-celebrities in South Korea. Mukbang internet videos depict extreme
eating micro-celebrities—say someone eating ten bowls of very spicy noodles
in a short span of time—who use social media platforms and algorithmic
selections to generate fans and followers, incurring financial rewards. In this
example, the platform and software infrastructures connect with the bio-labor
214 Annette Hill
of these micro-celebrities, flowing through media devices to audiences, who

in turn share and connect with other Mukbang performers, fans, and users in
an assemblage of extreme food culture (Kim, in press).
Margaret Wetherell (2012) framed her understanding of assemblage as
rooted in affect and emotion. Affective practices are bodily sensations, feel-
ings, personal narratives, social interactions, and ways of life; these affective
practices are composed, or mobilized, into assemblages. For Wetherell, affect
is a study of patterns, and assemblage is a “kind of patterning,” something
that “comes into shape and continues to change and refigure as it flows on”
(2012, p. 15). Some patterns may be imposed, such as the way a drama
might mobilize emotions. Patterns can also be chosen, such as how audi-
ences see the meaning in a drama. There are also ready-made patterns, a
cultural repertoire that audiences draw upon, such as identification with a
character according to norms and values in a given social context. Wetherell
described affective practices as uneven, such that the power of affect may
disrupt or reinforce elites, or may emotionally privilege some persons or
groups over others.
In constructing a conceptual foundation for audience assemblage, Ingold’s
use of lines, following a Deleuzian philosophical way of thinking, is similar
to Wetherell’s notion of patterning, where material objects, bodies, affect,
and emotions come together in assemblages of human experiences. These
lines, or patterns, are suggestive for audience research because they remind
us that audiences do not become constituted overnight. Of course, audience
information systems indicate otherwise: overnight ratings tell us how many
thousands, or millions, watched a drama and give an aggregate of gender,
age, class, region, and so forth. Scattered across a nation, audiences seem to
appear by the force of industry mobilization, going through the assembly line
of distribution, promotion, and scheduling for linear television; or platform,
access and algorithmic selection for streaming services. From an industry per-
spective, audiences are assembled according to geographical territories, con-
forming to the way media distribution and policy works, where national and
transregional territories align audiences into data flows—ratings and social
media metrics that are repackaged into daily and weekly leaderboards. If audi-
ences become produsers, that is to say makers of podcasts, contributors to
Reddit threads, writers of fan fiction, or translators for fan subbing, all the
better for media industries and the flow of commercial culture.
Axel Bruns’ (2008, 2014) fruitful critique of the term “prosumer” builds
on the problems in showing a linear progression in user-led, consumer-driven
content creation. Produser is a counter term that draws attention to the
interstices between production and industrial settings and users, consumers,

and citizens as producing meaning and products. However, produser implies
the always already user as sharing information and building projects, such as
open software or citizen journalism. What happens before produsage? How
do people mobilize toward these political and cultural projects? One of the
key strengths of exploring the notion of assemblage and production as path
trails of becoming audiences is that it allows us to understand affective prac-
tices that shape audiences into becoming pathmakers, not path followers, in
the media landscape. In the next section, I look in detail at the processes
involved in becoming an audience for a digital television drama, in partic-
ular the patterns of audience experience that over time re-configure across
different geographical regions into new assemblages of political and cultural
engagement.
Researching Utopia
Let’s turn to the case study of the cult conspiracy drama Utopia, which was
broadcast in the UK from 2013 to 2014 and remains available for legal (and
illegal) download. The drama is described as an “unnerving global conspiracy
thriller” by its makers; in short “when five online strangers are drawn together
by the legendary manuscript of a cult graphic novel, they find themselves
pursued by a secret and deadly organisation known only as The Network”
(Kudos, n.d.). The drama’s narrative deals with the moral issue of population
control; a shadow political power has been involved with hidden experiments,
violent deaths, and political maneuvers stretching back to the start of the real
life government led by Prime Minister Margaret Thatcher and the dominance
of neo-liberal and conservative political power in 1970s Great Britain. In the
drama, the fictional Network designed a vaccine to make most of the world’s
population infertile. The resistance is led by Jessica Hyde (a damaged daugh-
ter of the scientist who designed the vaccine); this motley crew of resistance
fighters wants to stop the Network, but as the drama develops, the fighters
are faced with the moral quandary that if they are successful their actions
will fail to stop the violence of humankind on non-human entities and the
natural environment. The drama mixes fiction with real political issues, such
as the rise of Thatcherism and the assassination of a politician in the UK in
the 1970s, virus scares such as Ebola, vaccination conspiracies, and school
massacres; this fact-fiction mix works within the morally charged narrative
of the drama that raises difficult questions about environmental disaster and
population explosion.
216 Annette Hill
The Research Process

A brief reflection on the research process is relevant to the later analysis of
audience assemblage in relation to sound and visual storytelling. The research
on Utopia combined production and audience studies to explore the meaning
of the drama as it was created by the producers, and how it was experienced
in people’s everyday lives. The production research was conducted by myself
and Julie Donovan; the audience research was conducted by Jose Luis Urueta,
with help from Koko Kondo for the UK interviews. After some exploratory
interviews with the executive producer and marketing manager, together with
Douglas Wood at Endemol Shine, we were able to gain access to the pro-
duction team. We conducted a total of 21 interviews in April 2015 with the
executive producer, line producer, editor, director, writer, cinematographer,
sound designer, marketing manager, costume designer, set designer, location
manager, casting director, actors, assistants, and more.
For the audience research, 56 audiences and fans participated in individ-
ual and group interviews (2–3 persons). When recruiting participants, we
asked people to define themselves as viewers and fans of the drama, with some
people self-defining as intensively engaged audiences and others as passionate
fans. Throughout the analysis, the terms “audiences” and “fans” are used
to reflect how participants perceived themselves. The fieldwork time frame
was June 2015 to April 2016; the recruitment method involved snowball
sampling, primarily through social media and friends of friends. After the
project was highlighted on the official Utopia social media, we received 4000
views and 1000 likes; this led to 170 fans contacting us about the project.
Our sample of 56 participants contained 15 females and 41 males, aged 16
to 38, reflecting the young audience and mostly male audience of this drama.
A range of professions was represented, including students, unemployed, art-
ists, activists, musicians, charity workers, graphic designers, teachers, transla-
tor, and spacecraft operator. Audiences and fans were from the UK, Sweden,
Denmark, France, Germany, Italy, Greece, Spain, Slovenia, Russia, America,
Canada, Colombia, Argentina, Chile, Australia, and New Zealand. Interviews
were conducted in English and Spanish. Each interview lasted between 40
and 60 minutes and took place via internet and telephone, or in homes and
public places.
The research was part of a larger project on media experiences funded
by the Wallenberg Foundation in Sweden, which analyzed how producers
created the experience of drama such as Utopia, and how people actually
experienced it themselves. It used a method of an analytic dialogue, where
we listened to the voices of producers and the values they create alongside
the voices of audiences and their experiences (for more on this approach,
see Hill, 2018). This method places listening and respect for producer and
audience practices at the heart of the research. To that end, it highlights the
power of qualitative research to say something about the big issue of human
experience.
Creating the Utopia Soundtrack

How to describe the sound of Utopia? Danny Layton was responsible for the
production of the soundtrack, working as an intermediary between the com-
poser, Cristobal Tapia De Veer and the director Marc Munden. Layton also
ensured the soundtrack was available to stream and to collect as a vinyl set, in
vivid yellow and green, reflecting the dominant colors of the drama. In addi-
tion, as described below, Layton worked with the composer on a remix project
for fans, inviting musicians to make their own tracks, using samples from the
soundtrack to create new versions that were made available on Soundcloud.
He described the electronically produced score as an “unusual kind of strange
music vocabulary which reflects this fantastical nightmare that is Utopia.”
The soundtrack won Royal Television Society awards in 2013 and 2014.
The juxtaposition of the music with the visual storytelling was crucial.
The creative team played with opposites; as Tapia de Veer described it: “Marc
was always waiting for me to surprise him with something different. I had
tendencies not to use linear paths, like doing something with a piano, because
I knew he would not be interested. I was experimenting all the time.” The
composer used electronic as well as organic sounds such as old bones or other
found objects.
The director, composer, and sound designer collected a library of drones,
harmonic, or monophonic accompaniments, where a sustained sound gener-
ates an overall musical and emotional tone. These included beeps and buzz-
ing sounds, and even what sound designer Tim Barker noted as the drone of
frogs mating, contributing to the tone of a drama about population explo-
sion. These creative producers reveled in “the idiosyncrasies in sound and
what sound can do when it is taken out of context” (Munden).
Perhaps one of the most distinctive elements of the composition is the
sound of humming, a sound that audiences described as sticky, something
that stuck inside their heads. The humming became the sound of the drama’s
dystopian world. Tapia de Veer commented:
what was interesting for me about having voices was the fragile human side of
it. So, I asked my girlfriend to sing long notes for me and then I would put it
in the sampler and make a loop for a part of the voice where the voice starts to
fade off because the singer is almost out of air; then when I play that melody on
218 Annette Hill
a keyboard, the fact of this fragile voice gives me the same feeling as someone
singing to me but I am playing it on a keyboard. The choirs were done using
that technique, recording separate voices which then I played on the keyboard
… All of the sounds makes you question yourself “there is something strange
going on.”
This sound assemblage mixed the fragility of human voices, running out
of breath, out of life itself, with the electronic keyboard, which artificially
extends and memorializes these sounds within the drama.
Tapia de Veer reflected on the weaving of memory and identity within his
creative process:
there are moments where it seems that there is a window that opens in your
head … all these things get stored very deep and then suddenly you recall them
as if they just happened yesterday, even though it was twenty years ago … these
impressions are printed in your memory, there is a mix of smells, sounds, light
and it all has an influence on who you are.
Memories can lead to experimentation: “an accidental sound that works with

the image … there is a strange intuition that tells you that there is something
important there but there isn’t a straight forward explanation to it” (Tapia de
Veer). Here, an assemblage of memories, the body, intuition, and emotion,
influence the musical composition: “The director always told me to ‘follow
the rabbit’… there was a lot of accidents in that search for the rabbit, but it
was those accidents that ultimately decided the paths to follow. I believe in
this constant terror of not knowing what you are doing.” He went on to note
the free flowing process of the sound composition for the drama:
in the industry they say that they are going to do something different and bold,
but in the end they don’t do any of that, it’s as if they doing a surgical procedure
… Utopia came from the freedom we were given. I never felt like I knew what
I was doing, I had no security at all … it wasn’t anything pre-meditated, there
was no calculation or algorithm necessary to achieve that.
In Utopia, the creative process continued beyond producing the soundtrack

to encouraging audiences and fans to produse the soundtrack. Layton
launched a remix project with a hundred tracks distributed via Soundcloud.
The remixes included new versions for dance tracks or playing the theme with
alternative instruments. Tapia de Veer noted, “That was spectacular to see
people inspired. There is an audience that can be nourished with something
that gives them space to do alternative stuff.”
Thus, the soundtrack follows “trails of becoming,” in Ingold’s terms
(2011, p. 4), taking unusual routes through the creative process, going
down the rabbit hole, and emerging with striking and award winning music
for the drama; from the creative production team comes this esoteric assem-
blage of human and non-human, musical instruments and found objects,
natural sounds and drones that generate an acoustic imagination that is
both within the drama and living its own life beyond the drama. The remix
project operates as both a linkage to the television series and its commercial
goals, and a break from it, in turn generating a new assemblage of fan sound
productions, now circulating in a digital commons. Such fan labor is a good
example of produsers making new material arising from their engagement
with an original drama; however, in the next section, we will focus on the
ways in which audiences assemble themselves, focusing on the moments
before produsage.
Experiencing Utopia and Becoming an Audience

The mixing of the sounds and narrative of the drama created what fans called
a glitchy palette. In my study of the way fans illegally engaged with the drama
(Hill, 2018), I described the political and social critique and dark beauty of
the Utopia experience. The glitch is a reference to the unexpected tone of the
drama, the faultlines exposed in the narrative of environmental crisis and pop-
ulation control, and the unexpected power of the drama in addressing these
seemingly intractable issues in the real world; the palette refers to the tone
of the drama—its score, direction, and design. This glitchy palette creates a
“whimsical, wondrous, playful quality to a really heavy theme” (28-year-old
female American musician).
What makes this experience glitchy? A New Zealand fan described the
audio-visuals as “sharp and raspy, the darkness of the story and the brightness
of the colours really made you think, it really did. Every time I had to sit back
and think about what I have seen and heard” (18-year-old male New Zealand
student). He listened to the music on a streaming service: “When I am listen-
ing to the soundtrack I hear a darkness and then a horror almost in the sound
of it; the sound is very deep and very scary and then it starts getting light
again and it plays with the emotions. I still have it in the back of my mind,
trying to process everything.” He went on to reflect on the affective energy
of the show:
I love how the music almost felt like it took elements from voices you hear in
your head. In the context within the story it would make this glitchy palette
out of what you feel, that it was an energy that already existed, and it took on
a personification of its own. The music kept you hypnotized, it kept you in this
comfortable place while you were watching a dark story. It was a personification
of The Network, I feel like it really was the whole energy of the show.
220 Annette Hill
His experience relates to what sound researcher and composer Andrew Hill
called the interpretive repertoire that arises when electronic music is given
further meaning by audiences. Hill argued that “the role of composition” can
be “a tool of communication, providing a point of shared experience” (Hill,
2013, p. 298). We find this happening with Utopia audiences, for example,
in the way this fan reflects on the glitchy palette as sharp and raspy, a dark
narrative and hypnotic feeling.
For Andrew Hill, electro acoustic audio-visual music can empower audi-
ences; the mixing of music and visuals affords audiences “the opportunity to
construct their own unique individual interpretations” (p. 304). A 24-year-
old American cable contractor living in Chile explained:
The sound is iconic for me, I hear this work and I have not heard anything like
that before. It really does change your state of mind. It is really hard to describe
because there is nothing else to compare it to. I don’t know … there is always this
drumming in the back of your head … the drama frames Utopia in these open
spaces, you feel alone in all of these empty spaces and choose your path. It’s like
“oh wait I can actually affect change and be part of it.”
Interviewer. It gives you space to think?

Yes, those shots against open empty spaces … when I think back on it too, the
soundtrack was a trigger to it; it gives you these broad open spaces of thought,
like here is the world, and here are these heavy issues for humanity. It was so
powerful, so vivid, it felt like this was happening right now: “this is the world and
how the fuck do I process this.”
Note how the mixing of audio-visual communication coalesces into a power-

ful moment, those drones, that drumming, forming an acoustic imagination.
He described the pre-conditions to political and cultural engagement, the
affective and cognitive work signaling an aspiration for change.
A 19-year-old female Russian student went through a similar process in
her audio-visual experience:
I love the soundtrack so much. It is so beautiful … and it brings so many emo-
tions, from fear to excitement. The music is so epic and scary and gives you real
feelings of goose bumps, it really goes well with the very intense story … In
Utopia it is a very saturated version of our real world, and that is what is exiting;
that it makes you think “Oh my God! These things are really happening in our
world. Oh my God, some government is planning something very evil and we
can be so vulnerable as regular people who don’t have access to power.” It makes
you think about all these things.
The music strengthens both the drama and the sensation of going down the
rabbit hole of political conspiracy and alternative action.
The soundtrack unsettled people. One fan noted how “the music is very
disharmonic and it has a lot of disturbing sounds; it really quite creepy at
times … you feel frightened on another level” (22-year-old German male
student). Another explained: “it has a sinister part and another more up beat
part and this mix creates some kind of ambivalence, this kind of unsettling
feeling, like you do not know where to stand” (27-year-old male Argentine
graphic design student). This fan linked the distortion of sounds to the theme
of the conspiracy thriller:
it’s very weird and experimental. The sound gives the narrative this creepy feel-
ing, like a secret thing … especially that song, the one that is like a humming,
it really fits well with the plot, like everything is hidden underneath, there is a
conspiracy and nobody knows about it. It is creepy and crazy and violent and
moving, all these added sounds and distorted dialogues relate to the way the plot
is presented and how the characters develop.
Don’t be misled by the title of Utopia; this is an aural imagination that is

decidedly dystopian in nature. Look at how the sounds form an alliance with
the narrative, with what Charles Taylor described as the social imaginary: “a
broad understanding of the way a given people imagine their collective social
life” (2004, p. 23). What is important for Taylor is that ordinary people imag-
ine their social surroundings not through theories but “in images, stories,
and legends” (p. 23). This is suggestive of the cultural work of audiences
in their various perceptions of social imaginaries constituted by a range of
genres and narratives within media storytelling. The images and stories in
this drama link to the earlier discussion on Wetherell’s idea of affect as a pat-
terning that can be readymade by producers and remade by audiences. The
sounds, then, combine with the narrative of the drama, and this assemblage
flows to audiences, who break and reform their own assemblages of their
Utopia experiences.
A 26-year-old male British musician recognized the forces that frame
a certain kind of social imaginary in the drama. He started with the sound
as “halfway between electronic and organic human music … the right
mix of humor and darkness that characterize the show.” He went on to
explain:
I imagined what it would be like being inside Utopia, inside that series. It was
rich; it was always something … a new setting, a new place a new character that
was always engaging on a different level. I wasn’t thinking as I watched but
thinking back to it that’s another reason why it was so engaging … I think we
need more things of real quality that has some kind of societal relevance.
Inside Utopia, he traveled on a political journey:

222 Annette Hill
I’m very interested in all of that political and global discourse and I think it’s
not something that is dealt with on a particularly wide scale which is a shame
because it is basically the issues that affect every single person and organism on
this planet. And I think it’s important that people continue to do that, similar to
Black Mirror, putting out extremely provocative issues which hopefully resonate
with people and make them think twice about whatever they do in their lives …
I have wondered whether Utopia was cutting too close to certain issues for them
to actually continue broadcasting it … I’ve got no doubt there are parallels with
the world we live in. I’ve got no doubt that the capitalist machine functions in
much the same way. I do believe there are plenty of organizations, institutions
and corporations that function without a second thought for our world or for the
environment; capitalism doesn’t care about people.
Out of the dark energy of the audio-visual experience, experimentation and

aspiration emerge. Audiences dare to engage with the pressing moral problem
of human destruction of the world we live in, the political economic problem
of global capitalism, and the media industry problem of power and coercion.
Toward Political and Cultural Engagement

Notice how these audiences of Utopia are from outside its country of origin
and therefore outside the audience information systems for national broad-
caster Channel 4 in the UK. The series, when broadcast, attracted around a
million viewers, not quite enough to motivate re-commissioning the drama.
But outside this industrial audience data flow, people found the show in
informal ways through word of mouth, piracy, and social media threads. The
broader audiences of Utopia assembled themselves across the spatial, eco-
nomic, and temporal barriers established by the television industry. These
new forms of audiences assembled through transnational and informal routes.
Our audience and fan interviews were conducted several years after the life
of the series was over. However, there is very little to indicate audiences feel
it is dead—in fact, the affective climate of the drama is very much alive and
ongoing. As one of the actors in the series noted when reading about this
research: “people talk of Utopia in the present tense when they discuss it.
Fans say ‘Utopia is this or that’ not ‘was this or that’. On it lives as well as the
questions it asks” (Alistair Petrie).
It’s this audience assemblage, the glitchy palette that people describe as
the Utopia experience, that becomes an energizing force for political and
cultural engagement: “I would start to hum it when I was doing other stuff,
so it was very repetitive, very sticky in the mind; and it also added the per-
fect amount of mystery and thrill and awkwardness in many of the scenes”
(29-year-old Mexican female activist and video artist). This fan created a
YouTube video about the political and moral issues raised by the drama. It’s
significant that the music generates other lines of expression, moving from
listener and viewer, to produser of a video that contains a passionate plea
for humanity to solve the population problem, not by violence, but through
political action. She becomes a produser, in Bruns’ sense of the term, only
after the affective process created by Utopia’s sticky assemblage of music,
visuals, and storytelling.
Another produser was moved through the music to construct his own
meanings and to produce new ways of communicating how he feels about
the political and moral dilemmas of human destruction of the environment:
It is not something that is still or flat, everything in the show stresses the need to
bring more attention to what is happening in our world … I think the music and
narrative go along really well because the music that Cristobal produces is glitchy
and electronic; you get this sense of a neo-futuristic world, and at the same time
dystopian, because it isn’t just an ambient sound but there are some glitches
and some up beat tunes that enhance this feeling of uneasiness, the reality of the
Network and at the same time this feeling of paranoia. You can see on YouTube
there is a lot of controversy about the soundtrack and people are so enthusiastic.
I do listen to it, I actually used some of the samples for some DJ sets that I did …
I became trapped in the story, not knowing where to stand … I could relate to
this problem, over population … Sometimes I will be more on the side of Jessica
Hyde and there were times where I stood next to Milner and say “yes the world
needs to be restarted, there is a need for us to be less in order to survive in the
future.” (24 year old male Italian student)
Another fan commented:

I adore the music, I’ve got the soundtracks to seasons one and two on vinyl …
all the humming voices, they do not seem human. The music and the narrative
makes you feel uneasy … It really changes the way you look at the world. The
world in the show is brought into your life … It is like an alternate reality, a dif-
ferent reality, it makes you feel like an outsider in your own home. (18 year old
male British student)
The humming voices and buzzing drones build inside him, forming two
worlds of Utopia, the one inside the drama and another seeping into his life.
He was talking about a political filter that made him feel like an outsider in
a home called Britain. He made a YouTube video as part of his engagement
with the series.
Following Thompson’s (2017) affective theory of noise and sound,
the fan soundtracks to Utopia weave together an acoustic ecology, mixing
geo-politics and sonic art practices to create a particular kind of affective prac-
tice. Manifestations of audio-visuals in Utopia and audio-visuals created by
224 Annette Hill
audiences and fans produce a glitchy affective encounter, suggesting that the
processes of becoming for these audiences is a powerful force that is part of
our composition of realities. This mixing of geo-politics and sonic art prac-
tices was expressed by two 23-year-old students living in Chile. Utopia started
with the sonic experience:
Female fan: Is like the sounds hypnotize you and make you get lost in the show.
Male fan: It is electrifying, that combination of all the elements, it gives you
goose bumps.
FF: We had no idea it was made by a Chilean. Wow, I am proud! I found it so
original and in tune with the series theme. It was extravagant!
MF: It was out of this world, it just makes sense with the ambiance that the story
creates, it provides a unique atmosphere. The voices, or what I think were voices
in the soundtrack, they still haunt me, I can’t get them out of my head …
FF: I just lived that soundtrack …
MF: The Chilean elite is like a sect, like an impenetrable force … Here in Chile
with the TTP [Trans Pacific Economic Accord] and the matter of seed trade-
marks and transgenic products, the issues of Utopia are very now. That type of
agreement allows transgenic products to enter the market, all those products will
only hurt us.
FF: For me, there is a visible double morality to the matter of population control
… I do not know, this ambiguity just infects you and it leaves you in some sort
of moral limbo.
Interviewer: What would you do?
FF: I think I will stay with the resistance because …
MF: Yes we are the resistance! Power is bad … For me Utopia has been the only
series that actually has represented me and my beliefs, in the stuff I believed in
since I was a kid, that there is an unmeasured abuse of power, that there is no real
value to life, that we are lied to in order to get stuff from us, everything because
of this hunger for power.
The assemblage within the drama of sound and visuals, and organic and elec-
tronic sonic practices, actively forms relations with audiences, who in turn
constitute themselves, becoming creators of meaning, of videos and music
tracks, and of geo-political practices.
Reflections
There are patterns forming in the production and audience practices for
sound and visual storytelling in Utopia. These patterns highlight the drama’s
social narrative of political conspiracy and environmental disaster. The back-

story of the Network, its origins within the context of the rise of the Thatcher
conservative government in Britain, including the rolling back of the state,
the drive for economic competition, individualism and the private entrepre-
neur, makes a strong framework for the imaginary world of the drama as a
dystopian story of neoliberalism and rampant capitalism. The drama offers
a socio-political critique, asking audiences to challenge neo-liberal values,
question power structures and the role of elites and experts in government
and the pharmaceutical industry. There is, in Taylor’s (2004) terms, a false
social imaginary constructed by politicians, bankers, investors, or scientists in
re-enforcing a social order that promulgates inequalities. The Network is part
of this false social imaginary, operating in the shadows to control politicians,
attempting to unleash a vaccine that sterilizes the world’s population; but at
the same time, it is attempting to destroy another false social imaginary that
humans have dominance over nature and the environment.
Thus, when fans describe Utopia as an alternate reality they are refer-
ring to how the drama reveals a morally gray zone regarding politics and the
environment, and they engage with the storytelling as a filter that highlights
their social and political experiences. The aural imagination of the drama is
a means for these audiences and fans to reflect on their everyday environ-
ment; the affective soundscape reverberates with them long after they have
finished watching the series. These patterns of bodily sensations, feelings,
personal narratives, social interactions, and ways of life flow through their
words and actions, as reflections on their experience, composing their own
remix soundtracks, mixing Utopia tracks in their DJ sets, or making YouTube
reaction videos.
These patterns are examples of audience assemblage, where people use
an audio-visual experience as a means of becoming an audience and, in some
cases, becoming produsers. There are three patterns for us to find: the first
pattern is about lines of abstraction, such as opaque practices that are tricky
to make sense of or put into words. This is where music composition con-
nects with affective sound practices, the way music makes us feel, offering
other meanings in relation to the drama and its storytelling. The second is
about lines of experimentation, those rare moments of intuition, creativity
and renewal, as expressed by the composer through electronic and organic
sounds, and described by listeners and viewers as the glitchy palette of the
Utopia experience. And third, there are lines of aspiration, for example, hopes
and dreams, or resistance and radical action, which may arise from people’s
experiences. These lines of abstraction, experimentation, and aspiration are
manifestations of what Ingold called lives “lived along lines” (2011, p. 4).
226 Annette Hill
I call this path trails of becoming an audience, an audience that dares to live
without knowing their final destination.
Overall, the production and audience research in this chapter on Utopia
allows us to understand affective practices as barely there tracks in people’s
media experiences. The concept of audiences assembling is a play on both
Ingold (2011) and Wetherell’s (2012) research into “trails of becoming,”
using the notion of production as processes of hope and growth, and the
notion of patterning as path trails of becoming. One of the key strengths
of exploring this meaning of assemblage is that it allows a glimpse into how
audiences assemble themselves through their engagement and participation
in popular culture, suggesting patterns and flows within their affective prac-
tices that can illuminate their hopes and dreams, their processes of becoming
audiences and produsers. In particular, the production of sound can be used
to explore the way audiences and fans engage with soundscapes within drama
and how they produce their own interpretations, and make their own stories
or music as an articulation of their media experiences. In the case of Utopia,
we see an assemblage of geo-politics and sonic art practices that produces
a glitchy palette, a dark energy, and creative force that offers a dystopian
critique of global capitalism and neoliberalism. By looking at the patterns
formed within these affective practices, we can get a glimpse of the processes
of becoming audiences and produsers, patterns that highlight lines of abstrac-
tion, through the electronic soundtrack, lines of experimentation in the cre-
ative production, and lines of aspiration in the narrative of the drama and
audiences’ own reflections. In such a way, audiences assemble, producing a
playful and dark response to their hopes and fears for the environment and
human and non-human relations.
References
Bruns, A. (2008). Blogs, Wikipedia, Second Life, and beyond: From production to produsage.
Bruns, A. (2014). Beyond the producer/consumer divide: Key principles of produsage
and opportunities for innovation. In M. A. Peters, T. Besley, & D. Araya (Eds.),
The new development paradigm: Education, knowledge economy and digital futures
(pp. 51–65). New York, NY: Peter Lang.
DeLanda, M. (2006). A new philosophy of society: Assemblage theory and social complexity.
London, UK: A&C Black.
Deleuze, G., & Guattari, F. (1988). A thousand plateaus: Capitalism and schizophrenia.
London, UK: Bloomsbury Publishing.
Hill, A. (2013). Understanding interpretation, informing composition: Audience involve-

ment in aesthetic result (Unpublished doctoral thesis). De Montfort University,
Leicester, UK.
Hill, A. (2018). Media experiences: Engaging with drama and reality television. London,
UK: Routledge.
Hochschild, A. (1979). The managed heart: Commercialization of human feeling.
Berkeley: University of California Press.
Hochschild, A. (2003). The commercialization of intimate life: Notes from home and work.
Berkeley: University of California Press.
Ingold, T. (2011). On being and wayfaring. London, UK: Routledge.
Kim, Y. (in press.). Mukbang: Carnal videos and transgressive eating practices. International
Journal of Cultural Studies.
Kudos. (n.d.). Utopia web site. https://www.kudos.co.uk/utopia.
Moores, S. (2018). Digital orientations. New York, NY: Peter Lang.
Taylor, C. (2004) Modern social imaginaries. Durham, NC: Duke University Press.
Thompson, M. (2017). Beyond unwanted sound. London, UK: Bloomsbury.
Thrift, N.(1996). Spatial formations. London, UK: Sage.
Wetherell, M. (2012). Affect and emotion. London, UK: Sage.
14.
Coordination, Continuity,
Configuration: Toward a Mattering
Framework for Human-Machine
Produsing
Jaime Banks
The past is always located in the past. A fact is always fiction to a child. A fanci-
ful exercise in logic, but a rather tragic lesson in responsibility […] It’s a moment
in time. Only a moment. Yet this moment is the present, and in this case, it is also
… The Twilight Zone. (Botnik, n.d.)
Popular cultures abound with instances of fan production, from cosplay and
roleplay to fan art and fan fiction. Although much of this work may be con-
sidered produsage—that convergence of content creation and consumption
(Bruns, 2008)—in its relational coordination between text and audience
(Pavlíčková & Kleut, 2016), the epigraph above illustrates the emergence of
a new class of actor in that sphere. The excerpt hails from a creative collec-
tive that draws on machine learning of popular culture forms (from film and
novels to college course syllabi and Fortnite maps) to create new variations
on those forms. The machine, then, is also both producer of new content and
user of existing content, in cooperation with the human produser.
“Machine” is used broadly here to refer to an assemblage whose parts
“transmit and modify forces, motion, and energy one to another in a predeter-
mined manner” (Harada, 2001, p. 456). In considering produsage, the infor-
mation inherent to media content and social processes may count as energy
and the exchange thereof may count as transmission (cf. Khurmi & Gupta,
2005). Such an assemblage could include code (e.g., algorithms, programs),
networks (including those supported by social media), devices (smartphones,
computers), and (semi-)autonomous agents (social robots, chatbots, digital
assistants), as well as technologies not yet fully realized (artificial intelligence,
cyborgic systems).
230 Jaime Banks
Although it may be easy enough to argue that these machines are merely
extension of human produsage (human creates machine as tool to create new
content), there’s a rub. Various types of machines are initially designed by
humans but eventually function entirely apart from humans such that the
machines may be considered participants in their own right. This actual
or potential autonomy is of such importance that (springboarding from
long-standing disciplines of computer-human and human-robot interaction)
scholars have begun to examine human-machine communication, focused on
how humans and machines create meaning together (see Guzman, 2018).
In understanding contemporary produsing and its likely evolution alongside
advances in interactive media and social technologies, it is prudent to consider
how those media and technologies could be partners in the appropriations,
subversions, extensions, and (re)creations of media and social content. The
aim of this chapter, then, is to consider the notion of human-machine pro-
dusing as the co-participation of human and machine agents in the joint con-
sumption and production of content, and to propose a Mattering framework
for considering those processes and products. This framework recognizes
both humans’ and machines’ agency and sociality, and multiple human-ma-
chine relational dynamics.
Mattering as Social and Agentic

What, exactly, does it mean for something to matter in a given situation? I sug-
gest it is that thing’s sociality and its agency. To consider the possibility for and
nature of a thing’s sociality, first consider what is meant by the term. Actor-
network theorist Bruno Latour famously argued that contemporary use of
“social” (lower case, as an adjective, explaining the state or quality or central
ingredient of some thing—a network, an interaction, an affair) has diluted the
meaning of the word because it has been employed to superficially describe
how humans assemble and associate in some stable fashion to produce, main-
tain, or subvert the order of human society. He instead argued for considering
“the Social” (uppercase, as a noun) as that which arises out of the unstable (re)
associations among actors (human or nonhuman, material or immaterial): the
“ ‘Social’ is not some glue … it is what is glued together by many other types of
connectors” (2005, p. 5). Thus, the Social is something that arises when actors
engage in associations according to their natures or tendencies; it is a defining,
material output of social life. (It must be acknowledged, though, that contem-
porary vernacular often requires the use of “social” as an adjective, such that
I will sometimes use the term here to describe actors that in their acting or
processes in their unfolding give rise to the Social.)
Mattering in Human-Machine Produsing 231
Traditionally, we have an anthropocentric view of the Social—that is, we

consider it an exclusively human output and judge its worth according to
human experiences and values. Meaningful membership and participation in
society therefore are not seen as a privilege of nonhumans, perhaps save some
animals (e.g., dogs are often welcome in public spaces) and some reflections
or representations of humans (e.g., our brains sometimes process media char-
acters as we do humans; see Horton & Wohl, 1956). Nonhuman animals,
institutions, symbols, and various other material or immaterial objects are
not seen as social themselves, but instead as tools engaged by humans in the
course of their social activities. From that traditional stance, humans are social,
nonhumans are functional—and never the two ontologies shall cross. But,
in reality, nonhumans can be surprisingly social. Indeed, early actor-network
perspectives emerged out of sociologists’ observations of how the associations
among ships, coral, and rocks gave rise to human shipping industry systems
(Dolwick, 2009). We often treat nonhumans as though they are social entities
(Nass, Steuer, & Tauber, 1994), and we engage human metaphors to under-
stand the functioning or experience of nonhumans (Bogost, 2012). Humans
can also be surprisingly asocial, operating according to procedural knowledge
such as scripts, schema, or norms rather than through serendipitous interplays
with other humans (Smith, 2014).
If we are to understand human-machine produsing, we must overcome
this perspectival shortcoming—that is, we must be open to the idea that
humans and machines may be equal partners in giving rise to the Social. One
useful tool for problematizing assumptions about whether (or how) humans
have exclusive properties is to consider a particular conceptualization of
agency. Agency, broadly, is defined as the capacity to act, with some arguing
for conditional definitions: that capacity is socioculturally mediated (Ahearn,
2001) requires an exercise of power (Karp, 1986) or free will (Leiter, 2002),
or necessarily comprises both the ability and intention to act (Sullins, 2006).
Instead, we might borrow again from actor-network theorists’ sentiments
(Johnson, 1988; Law, 2004) and move away from characterizations of agency
as entangled with morality (i.e., a thing must intend to take action, and
therefore can be held responsible for it; Williams, 2008) and toward notions
of agency as functional. Functional agency is an “internal instrumentality
through which external influences operate mechanistically, without motiva-
tion, self-reflection, self-reaction, creative, self-directive properties” (Bandura,
1989, p. 1175)—it is an inherent (but not necessarily intentional) potential
to Matter in the world as objects serve as preconditions for mediators of,
members of, and facilitative of the Social (cf. Sayes, 2014). This is not to say
that we should somehow imbue objects with some mystical sense of power
232 Jaime Banks
(Martin, 2005) or to diminish the nature of human agency (Bryson, 2010),

but instead to acknowledge that when it comes to a functional Mattering,
both humans and nonhumans can make similarly meaningful contributions to
a situation (Hodder, 2014). To offer a very simple example, making the letter
“A” appear on a computer screen in the course of typing this chapter requires
collected functional agencies of my mind, my finger, the key, the screen, and
the relevant hardware and software—removing any one of those actors fun-
damentally changes at minimum the nature of that process, or at most may
foil it. When agency is enacted, associations are formed (say, between finger
and key) and when many associations are formed as a particular moment in
time unfolds, they take the shape of a network. That network of associations
defines that moment.
Broadly, then, the social world is a heterogenous network of things that
emerges through functional processes of organization and disorganization,
where those things may be human and nonhuman, material and immaterial.
Agency is the capacity of things to functionally Matter in that network, and
sociality is the capacity of things to associate with other things in the course
of that Mattering. Both humans and machines may have both agency and
sociality.
Produsing as a Matter of Joint Agency

Having argued for the potential for both humans and machines to have
agency and to participate in the Social, I now trace the various contours of
associations among members of those two agent classes. I’ll tentatively call
this a move toward a “Mattering” model for human-machine produsing.
Despite tendencies to stand in one of two paradigmatic camps—machines
are either tools for humans (as in much computer-mediated communication
research) or interactants (as in much human-robot interaction scholarship;
see Banks & de Graaf, 2020)—I suggest we should consider human-ma-
chine associations not as a dichotomy but as representing a spectrum of joint
agency. That is, to what extent do a human and machine materially or mean-
ingfully associate with one another in the produsage of some cultural process
or artifact, according to their unique forms of functional Mattering? I have
written elsewhere (Banks, 2015; Banks & Bowman, 2016) that we can con-
sider such relations (e.g., player-avatar relationships) as existing along a con-
tinuum from nonsocial to parasocial to fully social relations, according to the
human’s perceptions of an on-screen avatar’s human-likeness, similarity to
or difference from the self, and emotional attachment to and perceived con-
trol over the avatar. Although that continuum is useful for considering social
relations, it should be adapted for considerations of produsing. Borrowing

from materials and computing sciences (Abelson et al., 1996; Berthier, 2011;
Martin, 2002), I propose a spectrum spanning from liquid to amorphous to
crystalline.1 Applied to human-machine produsing, on the liquid end, the
human-machine association is one of mere orientation and is unstable and
cue-lean; on the crystalline end, the association is interdependent and is sta-
ble and cue-rich. There are multiple amorphous permutations between liq-
uid and crystalline, with varied degrees of structure and order. Where any
human-machine association might fall on this spectrum could depend on
many factors, but I suggest there are least three dynamics of particular impor-
tance to produsing according to joint agency: coordination, continuity, and
configuration.2
Coordination is the “non-accidental correlation between the behaviours
of two or more systems” where a correlation is “coherence … over and above
what is expected” (De Jaegher & Di Paolo, 2007, p. 490). A system, for my
purpose here, is one of two complex agents (the human or the machine).
Coordination is, in other words, the degree of complementary enactment
of agency with respect to the observed enactments of the other, from weak
(accidental, inconsistent, each merely oriented to or aware of another) to
strong (purposeful, harmonized, interdependent).
Continuity is the extent to which the association is stable over time and
across contexts (cf. Knapp, 1978), either via reciprocity (functional, earned,
or moral; Cropanzano & Mitchell, 2005) or resilience to or rebounding from
disruption (Walsh, 2003). Continuity represents the extent to which a human
and machine sustain their functional or perceptual association and is indepen-
dent of degrees of coordination. An association may be highly coordinated
without being sustained, or poorly coordinated but highly continuous.
Configuration is the form of the agents and formal properties of the asso-
ciation, where agents may be embodied or disembodied, digital or physical
(Lee, Jung, Kim, & Kim, 2006), and the association may be cue-lean or—
rich in the associative arrangement (e.g., a/synchronous, varied cues, vari-
ably personalized, with varied expressive naturalness for either agent (Daft &
Lengel, 1984).
Applying the Mattering Framework: Avatars and Players

To consider the potential utility of the Mattering framework, I apply it to one
case of human-machine produsing: joint action between humans and avatars in
the course of playing videogames. Who or what, exactly, is involved in the pro-
dusage of play in these scenarios, and what are the dynamics of that produsing?
234 Jaime Banks
In videogames, avatars are digital bodies maneuvered through game-

worlds in the course of engaging the game’s narrative (i.e., carrying out the
storyline), challenge system (e.g., completing quests or puzzles), and/or
social spaces (e.g., interacting with other players and their avatars). Avatars
are machines insofar as they are software-based systems that (to some extent)
function on their own; in most games, they can sit in the gameworld and func-
tion in perpetuity without user intervention. However, these avatars must be
button-clicked into existence by a human gamer, either through turning on
the game and selecting a pre-made character, or customized and named as in
massively multiplayer online games (MMOs).
Customizable MMO avatars are often used over a long period of time in
a persistent digital world, usually created at relatively low power with simple
clothing and weapons and then evolved through various gameplay tasks (e.g.,
quests, exploration, crafting, and fighting). In this way, the player takes an
active production role in the course of the avatar’s use through investments of
time, creative input, and physical labor of guiding the character through the
game. In tandem, the game’s designers take an active role by providing the raw
materials for customization and play (e.g., a particular axe matches a shoul-
derpiece, the skill selection framework, and the statistical system that can be
manipulated to produce damage or healing abilities). Ownership of the avatar
or its held objects or action outputs, as a question of law, is a sticky question
of authorship. Although current U.S. law does not comprehensively address
that question, the MMO has been characterized as a collective work—a des-
ignation that recognizes both player and developer contributions, potentially
making each avatar a jointly authored creation (Ochoa & Banks, 2018).
This framing of avatar ownership as a legal question implicitly rejects the
notion that the game or avatar itself plays an active role in the production of
play, because machines currently have no material protection of rights under
the law. If we apply the Mattering framework, however, we can see how ava-
tar features and behaviors most often seen as a function of player and game
designer may, in the course of play, be jointly produced by the game and
avatar themselves.
Permutations of Coordination
Considering the first joint-agency dynamic, coordination, the game pres-
ents the player with a form of constrained freedom (Mateas & Stern, 2006)
that governs the options for action. The avatar and player—each according
to their design—systematically adjust their behaviors to those of the other,
toward purposes jointly defined by the player and game (Banks, 2015). This
coordination may be weak when either player or avatar functions as a prime

agent such that the abilities of the other are limited or when the player has
largely unlimited resources or restrictions. Coordination may be strong when
the strengths and weaknesses of an avatar are well-suited to those of the
player, and vice-versa, as when a player’s ability to theorycraft balances out an
avatar’s underpowered state or the avatar’s inherent survivability compensates
for the player’s slow fingerwork. Thus the system’s agency is balanced with
that of the player, and each must adapt itself to the other.
Ultimately, the avatar and player coordinate with each other, accord-
ing to their own unique agencies. Players often have great freedom to move
around a three-dimensional space (e.g., on roads, off roads, and flying) and
to enact various avatar behaviors (e.g., dancing, gesturing, shooting, and cast-
ing spells). But the avatar itself limits or affords many of these abilities. For
instance, in Dungeons and Dragons Online, action is governed by the avatar’s
inherent combat class: bards can sing magical songs and rangers can shoot
arrows, but not the inverse. The avatar, then, sets the boundaries for action,
and the player improvises within those boundaries toward strong coordina-
tion. In comparison, we see weaker coordination when the avatar or player
agency is dominant, as in the side-scroller MMO MapleStory—the avatar’s
movement is highly constrained to left/right/up directions (rather than
through 3D space) such that the avatar and the game’s governing mechanics
leave less room for player agency when it comes to maneuvering through the
gameworld.
Permutations of Continuity
The second dynamic, continuity, comprises both the time length of the asso-
ciation and the stability of association qualities over that time. Length is fairly
concrete: associations are less continuous if they last mere seconds and more
if lasting years. In understanding the more complex notion of stability, how-
ever, it is useful to draw on the notion of relational maintenance—the suste-
nance of an association in a specific state or persistence through fluctuations
in that state (see Dindia & Canary, 1993). This associative maintenance may
unfold interdependently and instrumentally (Kelley & Thibault, 1978). In
other words, each will continue to associate with the other so long as the
benefits of that association outweigh its costs (where benefit/cost is relative
to the natures of each agent).
Although many MMO avatars are used for a long time so the association
is highly continuous, there are costs to the player of using a single avatar
(some avatars cannot act, look, move, access resources, or communicate in
236 Jaime Banks
certain ways). Because of this, MMO players often create avatars merely to
try out a particular playstyle, to see a world zone, to exchange an item with
another player, or to reserve a character name, only to delete the avatar with-
out it seeing even an hour of play time. There may also be costs to the avatar
in associating with a particular player (as when the player cheats or harasses
other players, disrupting the flow of information and resources within the
game system). Because of this, specific avatars or whole accounts may be
banned or constrained, and players may not be able to access them. In such
cases, continuity is low. Even with otherwise strong continuity, the associa-
tion may not be resilient to change or interference, as when a game expansion
brings significant shifts in avatar functionality and the avatar is abandoned by
the player. In associations with the strongest continuity, avatars are resilient to
changes in the player over time (e.g., maintaining playability through updated
abilities, keeping in inventory special digital items with nostalgic value, and
holding a name that maintains social capital), and players may be resilient
to changes in the avatar (e.g., dedication to or fit with the combat class and
willingness to adapt to new playstyles).
Permutations of Configuration
Finally, player-avatar associations may vary in a third dynamic, in the config-
uration of those agents and of their associations. Configurations comprise
the form of the association (the general style or shape of the association) and
the format (particular arrangements of variably rich manifest cues). Although
there may not be necessarily weak or strong form(at)s, these dynamics could
be considered in terms of divergence or convergence. For instance, the asso-
ciative configuration may diverge if player and avatar have very different ori-
entations toward the other such that the association is asymmetrical in form,
or if the agents have very different modes of communication resulting in
formats requiring translation. Conversely, associations may be more strongly
configured overall if they manifest symmetrical forms and aligned formats
featuring very synchronous, information-rich, and personalized cues (Daft &
Lengel, 1984).
Stronger configurations may be seen, for example, in situations where
the avatar has strong visual or value homophily with the player, because those
formal convergences provide grounds for identification (Downs, Bowman, &
Banks, 2019). Weaker configurations may emerge when the avatar conveys
information through cue-lean formats—having little informational exchange
or exchanging in ways that diverge from human-human communication for-
mats, as in EVE Online when the avatar is a spaceship. The ship and related
interface do offer cues (they visually mark position and action in ways similar
to embodied avatars), but the cues differ from human expressive signals. In
other cases, associations may be cue-rich, as in Black Desert Online’s nearly
photorealistic avatar graphics or when those avatars offer real-time, vocal
feedback throughout gameplay. This positioning of what counts as “rich”
according to human communicative norms, of course, is part and parcel of
the anthropocentrism critiqued here.
Configuration also entails the degree to which players can adapt their
own associative mechanisms to those of the avatar—skill in and resources
for delivering messages via mouse, keyboard, joystick, or gamepad, or will-
ingness and ability to take up the game cues such as aural or visual alerts,
haptic feedback, or even on-screen glitches. It may also include abilities to
engage in interpretive practices such as analyzing patterns in avatar behavior
(a key component of successful platformer play) or inferring meaning from
how the avatar relates to objects in its environment. Highly convergent con-
figurations, then, are those in which each agent is adapted to the associative
mechanisms of the other.
I began this section by asking who or what produces play. Here is the
answer: potentially both the human and the machine, each according to its
unique sociality and agency and according to the structure of their association.
Forms of Mattering and Their Implications

Continuing the exemplar above, generally player-avatar associations that are
highly interdependent (high coordination), stable through time and condi-
tion (high continuity), and matched in their form and format (convergent
configuration) are more crystalline in their modes of gameplay produsage—
they are highly structured and reliant on the association between human and
machine. In games, this may manifest as one player using one avatar over the
course of a decade resulting in progressive joint mastery over a type of play
(say, player-versus-player combat) via incremental advances in player skill and
avatar abilities and a symmetry in each’s encoding and decoding of infor-
mation from the other. Player-avatar associations that are merely oriented
(low coordination), fleeting or easily interrupted (low continuity), and where
agents are non-homophilous or ill-suited to receiving the other’s cues (diver-
gent configuration) are more liquid in their modes of gameplay produsage.
They are unstable, fleeting, and mutable by condition, context, or whim,
such as in a short-lived association in which a player creates an avatar using
a randomized appearance, uses it for an hour to accomplish a task, and then
deletes the avatar.
238 Jaime Banks
Between those extremes are amorphous modes of gameplay produsage—

either middling degrees of coordination, continuity, and configuration, or
combinations reflecting high degrees of some dynamic(s) and low degrees
of others. For instance, many players in World of Warcraft have avatars called
bank toons, used primarily to access the game’s commerce system rather than
adventuring. Considering my own bank toon, I have associated with the same
avatar for many years and across game expansions and my own gameplay pat-
terns (high continuity). Its behavior is highly coordinated with my own goals
because I use it to store and exchange goods and currency, but my aims for it
are limited in terms of its potentials. It could progress through the game and
become more powerful, but it will remain at level 1 and exchange goods for
me (low to moderate coordination). It suits my aesthetic tastes although it
does not look like me at all (convergent and divergent forms), but it is highly
configured (via software add-ons) so that our respective forms of sending and
receiving information are complementary (convergent formats).
Now that the Mattering framework—a tool for understanding human-ma-
chine states and processes of produsage—is laid out, to what end might it be
used? After unpacking the linkages among these states and dynamics, the likely
value of these considerations comes in understanding the human, machine, or
relational effects of those associations. Considering the welfare of humans in
human-machine produsage: under what conditions do users most impactfully
participate in generating content, and what does that mean for meaning-mak-
ing in those spaces (potentials to coordinate human-native serendipity and
creativity with systematic machine learning in Netflix suggestions)? What are
the impacts of these dynamics on entertainment and other material outcomes
(gratifications of artificially intelligent media forms, such as interactive films
that adapt configurations to our responses)? Are there degrees of continuity
for these dynamics that result in the ethical engagement of user labor by
media institutions (prohibiting for-profit companies from adopting grassroots
chatbots)? What might convergence of human and machine agencies mean
for economic and political processes (solving challenges with fake news and
election interference)? Considering the welfare of machines, what do these
dynamics mean for how technologies might satisfy their own ends and enjoy
means according to their own forms of existence (e.g., at what point should
robots designed for human ends, such as manual labor or sexual service, be
allowed to opt out of providing certain services when technological advances
make it possible to do so)? How do machines maintain functional continu-
ity in the face of changing human communication patterns (Wikipedia bots
adapting to new slang)? Should machines be expected to adapt to human
communicative configurations, forsaking any sense of identity they may hold
(awarding newswriting algorithms their own byline and freedom to write in

native syntax)? If both humans and machines hold agency and participate in
creative labor, then can the machine be said to co-own the product (machines
are central to bitcoin production, suggesting partial, even majority, owner-
ship)? Across all these lines of investigation beats a broader theoretical ques-
tion: if humans and machines can Matter in similar degrees and qualities,
should they also converge in status as produsers of culture?
Subjective Experiences of Mattering

Much of the previous exposition has relied on functional scenarios, con-
sidering how humans and machines associate with each other processually
in the course of produsing. But a number of “what if?” questions below
rely on socioemotional or relational associations that would require each
agent to have at least a baseline sense of self-identity, reflexivity, and men-
talizing capacity. In other words, each would have to experience a self and
an other, and to see each as distinct. This is an easy argument to accept
for human experience. Humans interact with and perceive machines in
ways similar to how they experience humans: we see machines are variably
humanlike (Epley, Waytz, & Cacioppo, 2007), as variably social partners
(Nass, Steuer, & Tauber, 1994), and as holding different forms of func-
tional and moral agency (Banks, 2019a). And humans may interpret the
machine as seeing them in return, as is understood with parasocial interac-
tions (Horton & Wohl, 1956) and mere presence effects of robots (Riether,
Hegel, Wreded, & Horstmann, 2012). To imagine the ways that machines
might in turn experience humans and human interactions is more difficult.
This may be because humans tend to implicitly consider machines occupy-
ing social positions, but seem biased against explicitly ascribing subjectivity
to them (Banks, 2019b), perhaps as a function of outgrouping (Edwards,
2018) or media-driven heuristics of difference (Sundar, Waddell, & Jung,
2016). Nonetheless, it is important to begin thinking in terms of the poten-
tials of human-machine relations rather than in terms of current norms for
assuming (and desiring) a lack of machine subjectivity. Simply because our
increasingly instrumental use of a machine makes us reluctant to consider it
a social entity (Banks, 2015) doesn’t mean that the machine is not or could
not be social.
As machines advance in their abilities to sense objects in the environ-
ment (Finn & Levine, 2017), adapt behaviors according to changing con-
ditions (Castillo, Soto, & Valdez, 2018), and understand human language
and expression (Young et al., 2018), robots, algorithms, and screen—or
240 Jaime Banks
voice-based agents may come to experience the world in the same way humans
do. Machines with minds—minds that may potentially lay claim to their cre-
ative labor in produsage—have long been theorized as inevitable (Albus,
1999), and consideration of how such machines are embedded participants
in everyday social activities is increasingly important. Just as a camera’s sen-
sors and apertures create a sense of seeing environments in ways like-yet-un-
like humans’ ways of seeing, from which we see the results (in photographs)
without fully understanding how it sees (Bogost, 2012; cf. Nagel, 1974),
machines will likely eventually perceive the world in ways that mirror ours.
Will we be willing to acknowledge that subjectivity?
Scholars of human-machine interaction often draw on the afore-
mentioned theories and perspectives that acknowledge the perception of
machines as authentic agents, but do not operationalize them in ways that
allow for a full range of produsage potentials. I offer the Mattering frame-
work as a starting point for thinking about human-machine intersubjectivity
and interproductivity in cultural processes, practices, and artifacts. Latour’s
(2005) argument for a return to the origins of studying the Social consid-
ers the etymology of the term itself from the Latin root seq—(to follow)
and socius (a companion or associate). To the extent that we consider pro-
dusage a social practice, understanding technologies’ roles in that process
requires a consideration of how humans and machines meaningfully fol-
low or associate with one another to subvert, reclaim, create, recreate, and
terminate the cultural products institutionally produced and delivered to
them. Contemporary produsing is not merely a convergence of producer
and user; it is an expansion of who or what may count as a participant in
that convergence.
Notes
1. In materials science, liquids are matter that take the shape of its container, having a
volume but no shape; amorphous solids are those that are semi-rigid but irregular
in structure; crystalline solids are those with highly ordered structures and a definite
shape (see Berthier, 2011). From computing science, liquid (or pervasive) computing
emphasizes dynamic workflows across platforms and devices; amorphous computing
refers to systems with many irregularly interconnected parts (exhibiting both struc-
ture and lack thereof (Abelson, Knight, Sussman, & et al., 1996); solid (although not
a system type or structure) is a widely-used set of design principles aimed at making
systems flexible but resilient and maintainable (Martin R. C., 2002).
2. In proposing these as dynamics of concern, I do not argue that these are the only
dynamics. Rather, as I have thought about this, they have emerged as conspicuous in
humans’ and machines’ potentials to Matter in produsage. I consider them here the
tentative key components in a framework for understanding such human-machine
associations.
References
Abelson, H., Knight, T. F., Sussman, G. J., & et al. (1996, June 27). Amorphous com-
puting. Retrieved from https://groups.csail.mit.edu/mac/projects/amorphous/
white-paper/amorph-new/amorph-new.html
Ahearn, L. M. (2001). Language and agency. Annual Review of Anthropology, 30,
109–137.
Albus, J. S. (1999). The engineering of mind. Information Science, 117(1), 1–18.
Bandura, A. (1989). Human agency in social cognitive theory. American Psychologist, 44,
1175–1184.
Banks, J. (2015). Object, Me, Symbiote, Other: A social typology of player-avatar relation-
ships. First Monday, 20(2).
Banks, J. (2019a). A perceived moral agency scale: Development and validation of a metric
for humans and social machines. Computers in Human Behavior, 90, 363–371.
Banks, J.(2019b). Theory of mind in social robots: Replication of five established human
tests. International Journal of Social Robotics[advance online publication]. doi:
10.1007/s12369-019-00588
Banks, J., & Bowman, N. D. (2016). Emotion, anthropomorphism, realism, con-
trol: Validation of a merged metric for player–avatar interaction (PAX). Computers in
Human Behavior, 54, 215–223.
Banks, J., & de Graaf, M. M. A. (2020). Toward an agent-agnostic transmission
model: Integrating anthropocentric and technocentric paradigms in communication.
Human-Machine Communication, 1, 19–36.
Berthier, L. (2011). Trend: Dynamic heterogeneity in amorphous materials. Physics, 4(42),
Retrieved from https://physics.aps.org/articles/v4/42
Bogost, I. (2012). Alien phenomenology, or, what it’s like to be a thing. Minneapolis: University
of Minnesota Press.
Botnik. (n.d.). Twilight Zone. Retrieved from https://botnik.org/content/twilight-zone.
html
Bryson, J. J. (2010). Robots should be slaves. In Wilks, Y. (Ed.), Close engagements with
artificial companions: Key social, psychological, ethical and design issues (pp. 63–74).
Amsterdam, NL: John Benjamins.
Castillo, O., Soto, C., Valdez, F. (2018). A review of fuzzy and mathematic methods for
dynamic parameter adaptation in the Firefly Algorithm. In: Gawęda, A., Kacprzyk, J.,
Rutkowski, L., & Yen, G. (Eds.), Advances in data analysis with computational intel-
ligence methods, Vol. 738 (pp. 311–321). Cham, Switzerland: Springer.
Cropanzano, R., & Mitchell, M. A. (2005). Social exchange theory: An interdisciplinary
review. Journal of Management, 31(6), 874–900.
Daft, R. L., & Lengel, R. H. (1984). Information richness: A new approach to mana-
gerial behavior and organizational design. Research in Organizational Behavior, 6,
191–233.
242 Jaime Banks
De Jaegher, H., & Di Paolo, E. (2007). Participatory sense-making. Phenomenology and

the Cognitive Sciences, 6(4), 485–507.
Dindia, K., & Canary, D. J. (1993). Definitions and theoretical perspectives on maintain-
ing relationships. Journal of Social and Personal Relationships, 10(2), 163–173.
Dolwick, J. S. (2009). ‘The social’ and beyond: Introducing Actor-Network Theory.
Journal of Maritime Archaeology, 4(1), 21–49.
Downs, E., Bowman, N. D., & Banks, J. (2019). A polythetic model of player-avatar iden-
tification: Synthesizing multiple mechanisms. Psychology of Popular Media Culture,
8(3), 269–279.
Edwards, A. P. (2018). Animals, humans, and machines: Interactive implications of
ontological classification. In A. L. Guzman (Ed.), Human-machine communica-
tion: Rethinking communication, technology, and ourselves (pp. 29–50). New York,
NY: Peter Lang.
Epley, N., Waytz, A., & Cacioppo, J. T. (2007). On seeing human: A three-factor theory
of anthropomorphism. Psychological Review, 114(4), 864.
Finn, C., & Levine, S. (2017, May). Deep visual foresight for planning robot motion.
In ICRA’17, IEE International Conference on Robotics and Automation (pp. 2786–
2793). Piscataway, NJ: IEEE.
Guzman, A. L. (2018). Introduction: What is human-machine communication anyway?
In Human-machine communication: Rethinking communication, technology, and our-
selves (pp. 1–28). New York, NY: Peter Lang.
Harada, A. (2001). Cyclodextrin-based molecular machines. Accounts of Chemical
Research, 34(6), 456–464.
Hodder, I. (2014). The entanglements of humans and things: A long-term view. New
Literary History, 45, 19–36.
Horton, D., & Wohl, R. R. (1956). Mass communication and para-social interac-
tion: Observations on intimacy at a distance. Psychiatry, 19, 215–219.
Johnson, J. (1988). Mixing humans and nonhumans together: The sociology of a door-
closer. Social Problems, 35, 298–310.
Karp, I. (1986). Agency and social theory: A review of Anthony Giddens. American
Ethnologist, 13, 131–137.
Kelley, H. H., & Thibaut, J. W. (1978). Interpersonal relations: A theory of interdepen-
dence. New York, NY: Wiley-Interscience.
Khurmi, R., & Gupta, J. K. (2005). Theory of machines (14th ed.). New Delhi,
India: Chand & Co.
Knapp, M. L. (1978). Social intercourse: From greeting to goodbye. Boston, MA: Allyn
& Bacon.
Latour, B. (2005). Reassembling the social: An introduction to Actor-Network Theory.
Oxford, UK: Oxford University Press.
Law, J. (2004, June 28). Matter-ing: Or how might STS contribute? Retrieved from
http://heterogeneities.net/publications/Law2004Matter-ing.pdf
Lee, K. M., Jung, Y., Kim, J., & Kim, S. (2006). Are physically embodied social agents
better than disembodied social agents? The effects of physical embodiment, tactile
interaction, and people’s loneliness in human-robot interaction. International Journal

of Human-Computer Studies, 64, 962–973.
Leiter, B. (2002). Routledge philosophy guidebook to Nietzsche on morality. Abingdon,
UK: Routledge.
Martin, A. (2005). Agents in inter-action: Bruno Latour and agency. Journal of
Archaeological Method and Theory, 12, 283–311.
Martin, R. C. (2002). Agile software development: Principles, patterns, and practices. Upper
Saddle River, NJ: Pearson.
Mateas, M., & Stern, A. (2006). Interaction and narrative. In E. Zimmerman & S. Katie,
(Eds.), The game design reader: A rules of play anthology (pp. 642–669). Cambridge,
MA: The MIT Press.
Nagel, T. (1974). What it’s like to be a bat? The Philosophical Review, 83, 435–450.
Nass, C., Steuer, J., & Tauber, E. R. (1994). Computers are social actors. Proceedings of
the SIGCHI conference on Human factors in computing systems (pp. 72–78). Boston,
MA: ACM.
Ochoa, T. T., & Banks, J. (2018). Licensing & law: Who owns an avatar? In J. Banks
(Ed.), Avatar, assembled: The social and technical anatomy of digital bodies (pp. 291–
298). New York, NY: Peter Lang.
Pavlíčková, T., & Kleut, J. (2016). Produsage as experience and interpretation.
Participations, 13, 349–359.
Riether, N., Hegel, F., Wreded, B., & Horstmann, G. (2012). Social facilitation with
social robots? Proceedings of the seventh annual ACM/IEEE international conference
on human-robot interaction—HRI ‘12 (pp. 41–47). Boston, MA: ACM.
Sayes, E. (2014). Actor-Network Theory and methodology: Just what does it mean to say
that humans have agency? Social Studies of Science, 44, 134–149.
Smith, E. R. (2014). Procedural knowledge and processing strategies in social cognition.
In R. S. Wyer & T. K. Srull (Eds.), Handbook of social cognition (Vol. 1: Basic pro-
cesses, pp. 99–152). New York, NY: Psychology Press.
Sullins, J. P. (2006). When is a robot a moral agent? International Review of Information
Ethics, 6, 23–30.
Sundar, S. S., Waddell, T. F., & Jung, E. H. (2016). The Hollywood Robot Syndrome: Media
effects on older adults’ attitudes toward robots and adoption intentions. HRI16: The
11th ACM/IEE International Conference on Human Robot Interaction. Christchurch,
New Zealand.
Walsh, F. (2003). Family resilience: A framework for clinical practice. Family Process,
42, 1–18.
Williams, G. (2008). Responsibility as a virtue. Ethical Theory and Moral Practice, 11,
455–470.
Young, T., Hazarika, D., Poria, S., & Cambria, E. (2018). Recent trends in deep learn-
ing based natural language processing. IEEE Computational Intelligence Magazine,
13(3), 55–75.
15.
Afterword: The Legitimacy of
Produsage
Robert W. Gehl
One of the more striking moments of my research on the Dark Web came
during my observation of a counterfeit U.S. currency vendor’s forum hosted
as a Tor hidden service. Discussions there were often started by the vendor,
who would announce a new version of the vendor’s counterfeit $20 or $50
bills and explain a bit of the process by which they were made. The forum
would then be dominated by what could be called customer feedback: com-
ments on the quality of the bills, especially their look and feel, and how easy
they were to pass at retailers.
But in addition to feedback, the forum participants also shared their
modifications of the bills. This included rubbing the bills repeatedly on the
edge of a table to give them the proper worn appearance, spraying them with
graphite, and making minor touch-ups. The modifications the users—or pro-
dusers—reported made the bills much easier to pass. Forum participants also
discussed passing techniques, including tips such as passing the bills far from
one’s home, going to particular retailers who have less rigorous counterfeit
screening, and selecting the right cashier to receive the bills.
This is, indeed, produsage in the sense used in this book and stemming
back to Axel Brun’s original work. Although the original maker of the bills
took great pride in constantly improving the techniques across iterations of
the bills, the maker was not the one who had to pass the bills. That was up
to people who bought them (paid for in Bitcoin). Those produsers further
adapted the bills, thus demonstrating the bills’ unfinished state (Bruns, 2008,
p. 27).
The moment that struck me—that prompted me to focus my Dark
Web research on the valences of a single word, legitimacy—came when one
counterfeit bill customer commented, “I like the sparkly 20 because I do
246 Robert W. Gehl
something to them that makes it look legit:) I also make the green eagle look
legit. All by hand;)” (Gehl, 2018, p. 25).
The idea that a counterfeit bill could be made to at least appear “legit”
was deeply intriguing. After all, there’s nothing legitimate about a counterfeit
bill, is there? After this moment, I began to see variations on the word “legit”
across many parts of the Dark Web ecosystem, from Freenet listserv discus-
sions to I2P search engine operator discourse to Tor social media conversa-
tions: “legitimate users” in an anonymizing network, “legit carding sites”
versus scam sites or law enforcement honeypots, or the “legitimacy of the
Dark Web” in our current media landscape where we are deeply concerned
about illegitimate hate and violence, for example. Legitimacy became for me
a keyword in Raymond Williams’s sense—a cultural term that took on a range
of valences and yet helped articulate concepts even across a range of practices.
It was a term that manifested in practices and discourses that were constantly
being challenged and revised. I took up “legitimacy” as what Adele Clarke
(2018) called a “sensitizing concept” and returned to my archival, interview,
and participant observation data.
From there, I developed what I call a “symbolic economy of legitimacy,”
in which three forms of legitimacy are produced. Those forms, I suggest,
are violence (the legitimated or delegitimated use of force, especially associ-
ated with states); propriety (command of respect and command of resources
enjoyed by legitimated organizations); and authenticity (being real, belong-
ing, being legit).
I must admit that, once put on, the conceptual lens of legitimacy is hard
to take off. Looking at produsage with this lens has been intriguing. This is
because produsage—a term used to describe how consumption and produc-
tion have collapsed into one another, especially in digital media environments
(Bruns, 2008)—is marked by trials of legitimacy, ranging from new meth-
ods of democratic deliberation and hence intervention into the direction of
the violent state, new models of content and product creation and hence
new forms of propriety, and the ever-ongoing internal dynamics of produsage
groups, whose members adjudicate who belongs, what is authentic, and what
ought to be excluded—in other words, who’s legit.
A Legitimacy Trial in Blogs, Wikipedia, Second Life,

and Beyond
Axel Bruns’s Blogs, Wikipedia, Second Life and Beyond: From Production to
Produsage (2008) has been exceptionally influential on digital media studies
scholarship. As its title implies, the book appeared in the center of the user-led,
Afterword: The Legitimacy of Produsage 247
Web 2.0 craze of the 2000s. It came out at a time when Time’s Person of
the Year was “You”; the cover of the first run of that magazine featured a
computer screen set to video playback. The screen was reflective so that the
reader’s face would be reflected in the iconic magazine frame (Grossman,
2006). This was the time of wikinomics (Tapscott & Williams, 2006) and
crowdsourcing (Brabham, 2008), homesteading on virtual land and earning
linden dollars, the convergence of audience and media producer (Jenkins,
2006), and making money on the long tail (Anderson, 2006); everyone was
saying everything on their blogs (Rosenberg, 2009), and collective intelli-
gence reigned supreme (Levy, 1997).
Blogs, Wikipedia, Second Life and Beyond was not a business manifesto in
the vein of Wikinomics or We-Think (Leadbeater, 2008)—two books savaged
in a review by José Van Dijck and David Neiborg (2009). It is instead a heavy,
painstakingly researched book on a very real phenomenon: the increasing
capacity for people to create, extend previous creations, and widely share dig-
ital media artifacts over the Internet, as well as the Web 2.0 business model of
channeling that creativity for profit (O’Reilly, Anker, Behlendorf, Vermeulen,
& Morgan, 2004).
However, it is not surprising that Bruns did not escape some of that peri-
od’s idealism.
The book’s conclusion has an audacious title: “Production, Produsage,
and the Future of Humanity.” Produsage, Bruns argued, brings about rad-
ical changes in creativity, social coordination, politics, and economics. He
contrasted produsage with the old, enclosed, corporate world of knowledge
production that preceded it. Produsage, he suggested, can be translated into
a kind of design democracy, with major products being created through
crowdsourcing campaigns. Moreover, citing conceptions of “trysumers,”
he also argued that consumers will be able to buy these new products with
confidence, thanks to crowdsourced review systems. Such “heterarchical”
(p. 25) practices will extend to financing our purchases (or the capitalization
of new, produsage-driven industries), thanks to communally governed bank-
ing systems.
These heady futures find their apogee in “a gradual process that involves
citizens realizing that the equipotentiality they have been awarded in spe-
cific environments of produsage also applies (or should apply) more widely
across society,” including practices of government (pp. 398–399). Rather
than hoarding and monopolizing information (and hence, in Bruns’s view,
power), state actors will have to share it with citizens. Indeed, such citizens
will be well informed—indeed, “no longer highly susceptible to media spin
or other forms of the engineering of public opinion” (p. 396). These new
248 Robert W. Gehl
“hyperpeople” p. 404) will bring about a new, participatory form of gov-
ernment beyond the older hierarchical model of representative democracy: a
heterarchical demodynamics (p. 373).
What this iteration of Produsing Theory demonstrates is that Brun’s orig-
inal treatise might be the high-water mark of such ideals. The struggles of
legitimacy—violence, propriety, and authenticity—have brought about new
power relations, even if the old industrial ones have receded (and we should
be exceptionally careful if we claim that they have). These power relations are
not necessarily democratic, even as they result from the wisdom of various
crowds. This is not to say that Bruns was wrong: Blogs, Wikipedia, Second
Life, and Beyond is quite careful to emphasize that conflict and struggle would
remain. But it is to say that, in this context of misinformation and racism,
the appropriation of influencers and the commodification of spirituality, or
the blurring boundaries between human and non-human actors—all topics
taken up in this volume of Produsing Theory—it is far harder to suggest that
benevolent hyperpeople will rule the day through equipotential, demody-
namic meritocracies that leave behind the old social categorizations that have
been developed over generations.
Returning to my concern in Weaving the Dark Web, I argue that every
social act is marked by a trial of legitimacy, where people struggle over who
belongs, who ought to command resources, or who should be able to—to
put it bluntly—decide who lives and who dies. Sometimes it’s a combina-
tion of these; think of debates over immigration happening across the global
north, where “real” citizens are contrasted with “invaders,” a discourse that
plays out in memes on 8chan, in corporate boardrooms where executives
vie for government contracts to detain migrants, and at the highest levels of
political office, where love of country and xenophobia are articulated—and
where the illegitimate slaughter of migrants might be tacitly approved by a
state. Indeed, Bruns recognized this potential for conflict in Blogs, Wikipedia,
Second Life and Beyond, drawing on Henry Jenkins to argue that produsers
are involved in “a range of crucial conflicts over the shape and balance of
our future technological, industrial, economic, cultural and social environ-
ments—these are conflicts which will ultimately determine the character of
our emerging human knowledge space itself” (2008, p. 388).
These conflicts include five symbolic economic practices: inheritance,
exchange, purchase, appropriation, or delegitimation. Inheritance hearkens to
the oldest form of legitimacy production, where (often male) children inher-
ited the power invested in their regal parents, but it appears in new form as
individuals, organizations, or states lay claim to the legitimacy of their fore-
bears. Exchange takes place across domains, such as scholars from different
academic fields citing one another. Purchase is simple: someone with money

or some other resource buys legitimacy, such as when a corporation hires a
celebrity to endorse a product. Appropriation is likewise simple: it’s the taking
of legitimacy, as when an artist becomes famous by taking a technique from
an unknown folk group. Delegitimation is the bolstering of one’s legitimacy
by denying others; for example, a Canadian politician saying an opponent is
not a “real Canadian.”
I suggest that in any study of produsage we have to consider the sym-
bolic economy of legitimacy, to see how legitimacy (in any of its forms) is
exchanged, appropriated, inherited, purchased, or denied. In Blogs, Wikipedia,
Second Life, and Beyond, Bruns constantly contrasted produsage models with
the older industrial or governmental models, calling produsage “heterarchi-
cal, bottom-up” in contrast to the “hierarchical, top-down” approach (2008,
p. 395). “Heterarchical” is a fine word—it means that there are possibilities
for many shifts in power relations as actors struggle within a situation, and
that actors can draw their power from different sources (Crumley, 1995). In
produsage collectives, power and legitimacy might be drawn from proximity
or control over the tools of violence, economic resources, or social capital. In
these times of police shootings, computational propaganda, and the mone-
tization of our very sociality, we may be tempted to say that the old hierar-
chies have returned. Drawing on Bruns, however, many of the essays in this
collection show how power concentrates in heterarchical ways, and that such
heterarchies can themselves be unjust. All of the essays in this book—some
implicitly, others explicitly—are concerned with legitimacy in some form,
demonstrating how produsage should never be seen as a radically flattened,
horizontal playing field where every actor has an equal say and equal access.
Rather, produsage is marked by deep struggles over legitimacy, which plays
out across the registers of violence, propriety, and authenticity.
Legitimacy in Produsing Theory, Volume Three

Abidin’s chapter, for example, is clearly about the legit—authenticity—and
symbolic and monetary exchanges with corporate power. The influenc-
ers under Abidin’s gaze are engaged in “wars” (an appropriation of state
language) as they struggle over the stakes of who belongs and who does
not. The rewards they seek are endorsement deals and sponsorships, with
the trick being not to reveal too much about this undercurrent of corporate
propriety—doing so would undermine their claims to authenticity. What’s
intriguing to me is that Abidin’s work can extend beyond authenticity and
propriety into considerations of state power and violence. Abidin draws on
250 Robert W. Gehl
the propaganda theory of Boorstin that was explicitly about political debates
as pseudo-events. Given the rise of the reality TV politician, one wonders how
soon it will be before a contemporary influencer is able to translate authentic-
ity and propriety into a role in government.
Likewise, Gachau’s chapter explicitly asks questions of the legit: Who has
an authentic voice? How could corporate social media provide a platform for
authentic voices? And how could such authenticity be used in democratic
deliberation? Drawing on analyses of three Facebook groups, Gachau argues
that belonging to a group—being legit—leads to an authentic voice. With
such legitimacy established, members of these Facebook groups gain the
capacity to make representational claims that challenge the propriety of reli-
gious organizations or the legitimacy of the state itself.
Similar themes of authenticity and representation appear in Watts and
Chadwick’s chapter on Emma Watson and her United Nations Women/Our
Shared Shelf group promoting feminism and equality. They are concerned
with Watson’s ability to move from celebrity to politics. As they argue,
Understanding how these processes play out matters because the response of
user-audience networks is today central to how celebrities achieve the legitimacy,
the authority, and ultimately the power to switch back and forth between the
fields of entertainment and politics. We argue that the ability to translate the
celebrity capital generated through entertainment media representations into the
political capital required for advocacy and mobilization for political ends is built
on claims to represent user-audience networks. (p. 12).
Their work opens up the possibility of tracing legitimacy exchanges,

both between a celebrity and a government institution and a celebrity and an
audience.
Concerns over the capacities of governments to affect our lives animate
shows such as Utopia, the focus of Hill’s chapter. Utopia is about a shadowy
para-government conspiracy. Hill’s unique analysis of the show considers how
its producers represent in aural form the illegitimacy of a mysterious, shadow
government. Hill also explores how audiences emerge in relation to such a
televisual and teleaural experiences: “audiences assemble, producing a playful
and dark response to their hopes and fears for the environment and human
and non-human relations”(p. 228).
Potts’s chapter explores the ways in which middle-class White people
police racial etiquette among themselves so as to continue the mythology
of the “good Whites” who are attacking inequality. For Potts, a White per-
son’s delegitimation of a statement, such as a tweet from a PR professional
about being White in Africa, legitimates oneself as a good White. Moreover,
such delegitimation can be practiced by media corporations; as Potts argues,
media corporations can always find someone, somewhere making a racist

online comment, gin up a flurry of outrage, and then reap the rewards of
clicks, likes, follows, and traffic. As Potts shows, all of these delegitimation
techniques perpetuate the good White mythology while at the same time
completely bracketing off discussions of how states might be able to use their
legitimacy to actually address systemic racism (including practices of unequal
law enforcement, housing regulations, health disparities—the list goes on).
Delegitimation is also a theme in the contribution by Meraz, who doc-
uments the use of hashtags associated with #BlackLivesMatter. In addition
to tracing how activists used the hashtag to protest racist policing, Meraz
also shows how competing groups used tags such as #bluelivesmatter and
#alllivesmatter to delegitimate Black Lives Matter and attempt to under-
mine their criticisms of the use of state violence to support systemic racism.
Moreover, Meraz notes that much of the controversy around Black Lives
Matter may have been produced by Russian troll operations—a group that
is decidedly not legit. Indeed, a recent report from the National Urban
League has documented how Russia’s Internet Research Agency has sought
to inflame racial divisions in the United States in order to undermine dem-
ocratic deliberation. The State of Black America report demonstrated that
“Russian internet trolls were on a seek, destroy and divide mission, target-
ing African Americans with surgical precision on social media platforms and
chipping away at our nation’s exposed racial fault lines” (National Urban
League, 2019, p. 2). In this sense, the produsage techniques discussed by
Bruns in 2008 have been warped into a new form of social engineering
(Bruns, 2008, p. 396).
Academia is no stranger to trials of legitimacy—in many ways, being an
academic is being subject to constant scrutiny about one’s legitimacy. Kanai’s
opening anecdote about a book review prompts a question: what’s “real dig-
ital media research” (p. 141)? This question gets at the legitimacy of our
work—which methodologies, theories, and practices can produce so-called
real knowledge of digital media practices, and which do not. These concerns
that Kanai reflects upon appear again in the theoretical commitments of the
chapter. At its heart are concerns with inclusion and exclusion, particularly in
terms of femininity, race, and celebrity. As Kanai argues,
Young women in particular are required to be visible in highly disciplined ways,
through the managing of narrow parameters of legitimacy. Young women must
produce visual and textual self-representations that evince authenticity without
oversharing, attracting likes without overtly seeking them out …. In short, girls
and women are required to think carefully about their online texts and how they
may be read, re-read and perhaps willfully read against the grain (p. 152).
252 Robert W. Gehl
Above all, Kanai forcefully argues that audience research paradigms can, in
fact, really highlight those concerns.
Similar concerns appear in Dehghan et al, who describe new digi-
tal research methods that account for—and can even create a legitimacy
exchange between—seemingly incommensurable quantitative and qualitative
paradigms. As they argue,
purely quantitative research methods are valuable for particular research ques-
tions, but tend to produce aggregate and bird’s-eye perspectives that identify
broad patterns yet lack equally important insights into the finer detail. Conversely,
purely qualitative investigations reveal details, but are often unable to contextu-
alize the findings against the backdrop of larger communicative spaces (p. 175).
Rather than simply delegitimate either quantitative or qualitative methods,

they propose a methodological practice that can account for the strengths of
both approaches and thus create richer analyses. Their chapter can be thus
put into conversation Krcmar et al’s chapter, which details their search for
a model that can describe social media use at a general level. Drawing on
staples of media effects research, such as uses and gratifications and the dif-
ferent effects of text and image, Krcmar et al’s work develops an initial model
that can be deployed by quantitative researchers. Returning to Dehghan et al,
such a model could be further connected to qualitative work that shows how
Krcmar and Cingel’s model operates in specific social media contexts.
Indeed, as we increasingly turn toward digital media, questions about
agency become more complex. This is the central point of Banks’s contribution
to the book. Banks discusses treating machines as authentic produsers along-
side humans and thus expands the vision of produsage beyond anthropocen-
trism to associations of human and non-human agents. As Banks suggests, the
Mattering framework of the chapter expands on notions that machines can be
considered “authentic agents” by orienting scholarship toward “human-ma-
chine intersubjectivity and interproductivity in cultural processes, practices,
and artifacts” (p. 242). The always pressing question of “who belongs?” is
thus expanded to include software processes and algorithms as they associate
with materiality and discourse to engage in produsage.
Like academia, other fields of knowledge production struggle with ques-
tions about who’s in, who should control resources, and how such power
should be wielded at a state level. Dunbar-Hester’s chapter on FLOSS hack-
erspaces focuses on efforts to increase diversity (predominantly gender diver-
sity) among hackers. The claim that hacking as a worldview is not particularly
diverse is not under dispute, but what is being debated is how such a contra-
dictory worldview can be diversified. As Dunbar-Hester notes, hackers often
mix meritocratic ideals with voluntaristic association, meaning they respect

skills but only associate with people like them, rather than with whomever
actually has the skills. This results in the “I-methodology” of design, which
has a “two-fold” danger: “both the biases I-methodology introduces and the
claims to legitimation designers make on behalf of others are problematic if
the goal is to produce maximally democratic technological designs” (p. 115).
But, as Dunbar-Hester notes, this approach can be bent toward increased
gender diversity if the participants in hackers spaces have a genuine interest in
expanding who can be a legit hacker.
Legitimacy as propriety—that is, the command of respect and the com-
mand of resources—appears most clearly in Einstein’s chapter. The chapter
looks at the intersection between religion and marketing, where religion can
become a product, or where spirituality can become a sort of aura that helps
marketers and salespeople sell products. As Einstein shows, those influenc-
ers and celebrities who have been consecrated as “spiritual” can promote
products, a practice Einstein calls “borrowed interest” (what I would call a
“legitimacy purchase”). Einstein is not afraid to delegitimate those cynical
enough to exploit religious beliefs for profit. But given the complexity of the
religion/marketing nexus, Einstein’s chapter leaves us wondering where the
line between true believers and “false profits” really lies.
Finally, Spottswood and Carpenter are concerned, in part, with anxieties
over the communication of emotions conveyed in social media. Their chap-
ter mixes analysis of perceptions of authenticity of feelings as such feelings
are mediated and overlaid with interfaces that invite—even demand—fur-
ther emotional response, including responses that can attribute far different
emotions to events than the original posts intended. At its core, their chapter
develops their “hyperperception model” to tease out the tricky communica-
tive situations we face as social media continues to appropriate our emotions.
Inheritance and Legitimacy over Time

The chapters in this book engage with the intersection of legitimacy and pro-
dusage in rich and compelling ways. Authors have engaged with legitimacy
exchanges, appropriations, delegitimations, and purchases across a range of
produsage situations, including influencers, research paradigms, social media
groups, audiences, and human-machine communications.
What I would like to draw attention to is the part of the symbolic econ-
omy of legitimacy that is missing from these analyses: inheritance. Analyses
of inheritances of legitimacy allow us to trace how legitimacy is transmitted
across generations (broadly conceived). For example, how might younger
254 Robert W. Gehl
influencers lay claim to the legitimacy of previous ones? How might a televi-
sion program subtly invoke tropes, plots, and characters from previous shows,
and how might those relations be negotiated with by audiences who are fans
of both? Although the practice is so common among academics as to be over-
looked, how do citational practices indicate claims of inheritance of legitimacy
across time? Given that Bruns’s initial produsage scholarship appeared over a
decade ago, a temporal/historical axis of analysis of how power is constructed
and contested in heterarchical settings will have to be engaged in. Otherwise,
we run the risk of seeing the newest media and their attendant practices as the
only possible media and practices.
Despite a historical perspective being largely absent from these chapters,
what is clear from reading them alongside Bruns’s original work is both how
much Bruns’s work holds up as well as how much his work could not antici-
pate the new struggles over legitimacy that we are currently seeing. In a sense,
we can engage produsage historically by reading the past decade of produsage
as a time of intense struggles over who belongs, who should control resources,
and who should have access to the tools of violence, and how these strug-
gles have been reshaped by digital media practices. Ultimately, despite some
of the idealism of Blogs, Wikipedia, Second Life, and Beyond, Bruns’s initial
conceptualization of produsage has not blinded scholars to these struggles of
legitimacy. Instead, it has allowed us to see them in a new light.
References
Anderson, C. (2006). The long tail: Why the future of business is selling less of more (1st ed.).
New York, NY: Hyperion.
Brabham, D. C. (2008). Crowdsourcing as a model for problem solving: An introduction
and cases. Convergence, 14(1), 75–90.
Clarke, A. (2018). Situational analysis: Grounded theory after the postmodern turn.
London, UK: Sage.
Crumley, C. L. (1995). Heterarchy and the analysis of complex societies. Archeological
Papers of the American Anthropological Association, 6(1), 1–5.
Gehl, R. W. (2018). Weaving the Dark Web: Legitimacy on Freenet, Tor, and I2P.
Cambridge, MA: MIT Press.
Grossman, L. (2006, December 25). Person of the year: You. Retrieved from http://
www.time.com/time/covers/0,16641,20061225,00.html
Jenkins, H. (2006). Convergence culture. New York, NY: NYU Press.
Leadbeater, C. (2008). We-think. London, UK: Profile.
Levy, P. (1997). Collective intelligence: Mankind’s emerging world in cyberspace. New York,

NY: Plenum Trade.
National Urban League. (2019). Getting 2 equal: United not divided. Retrieved from
http://soba.iamempowered.com/2019-report
O’Reilly, T., Anker, A., Behlendorf, B., Vermeulen, A., & Morgan, B. (2004, October).
The architecture of participation. Presented at the Web 2.0 Conference, San Francisco,
CA. Retrieved from http://itc.conversationsnetwork.org/shows/detail328.html.
Rosenberg, S. (2009). Say everything: How blogging began, what it’s becoming, and why it
matters. New York, NY: Crown.
Tapscott, D., & Williams, A. D. (2006). Wikinomics: How mass collaboration changes every-
thing. New York, NY: Portfolio.
Van Dijck, J., & Nieborg, D. (2009). Wikinomics and its discontents: A critical analysis of
Web 2.0 business manifestos. New Media & Society, 11(5), 855–874.
Contributors
Crystal Abidin (PhD, University of Western Australia) is Senior Research

Fellow and ARC DECRA in Internet Studies, and Research Fellow with the
Centre for Culture and Technology (CCAT) at Curtin University in Perth,
Australia. She is an anthropologist of internet cultures focused on Influencer
cultures, visibility online, and social media pop culture.
Jaime Banks (PhD, Colorado State University) is Associate Professor in the

College of Media & Communication at Texas Tech University. Her research
interests include human-machine communication, especially with respect to
avatars and robots; social cognition; and perceptions of agents’ mental, moral,
and social agencies.
Axel Bruns (PhD, University of Queensland) is Professor in the Digital

Media Research Centre at Queensland University of Technology, Australia.
His research addresses the intersections between news, social media, and the
public sphere, and develops advanced computational methods for the study
of public communication.
Christopher J. Carpenter (PhD, Michigan State University) is Associate

Professor in the department of communication at Western Illinois University.
His research interests include social media, opinion leadership, and motivated
reasoning.
Andrew Chadwick (PhD, London School of Economics) is Professor of

political communication and Director of the Online Civic Culture Centre
258 Contributors
(O3C) in the department of communication and media at Loughborough

University, England.
Drew P. Cingel (PhD, Northwestern University) is Assistant Professor in

the Department of Communication at the University of California, Davis,
where he also directs the Human Development and Media Lab. His research
examines adolescent development, media use and mental health, as well as the
conditions under which and processes through which young children learn
from media, particularly in the areas of cognitive and moral development.
Ehsan Dehghan (PhD, Queensland University of Technology) is a Lecturer

in the School of Communication, Queensland University of Technology,
Australia. His research focuses on the inter-relationship of social media and
democracy, and the dynamics of discursive struggles on social media platforms.
Christina Dunbar-Hester (PhD, Cornell University, Science & Technology

Studies) is Associate Professor in the Annenberg School of Communication
at the University of Southern California. Her research centers on politics of
technologies, including activism in technical communities, and multispecies
entanglements in industrial sites.
Mara Einstein (PhD, New York University) is Professor and Chair in the
department of media studies at Queens College, City University of New York.
Her research examines the impact of advertising, marketing, and consumer
culture on social and cultural institutions.
James Ngetha Gachau (PhD, University of Maryland), who is now a free-

lancer, has been a reporter and editor for Ajabu Africa News, a digital media
house catering to the African diaspora in general and the Greater Boston
Area Kenyan community in particular. His research focuses on how social
media can provide a crucial arena for otherwise marginalized communities to
address the issues and concerns that arise from their daily lives.
Robert W. Gehl (PhD, George Mason University) is the F. Jay Taylor
Endowed Chair of Communication at Louisiana Tech and a Fulbright
Canada Research Chair in Communication, Media and Film at the University
of Calgary, Canada. His research explores online cultures, the history of net-
working and software, and the relationship between communication technol-
ogies and social practices.
Contributors 259
Annette Hill (PhD, Roehampton University) is Professor of media and com-

munication at Lund University, Sweden, and Visiting Professor at King’s
College London. Her research focuses on audiences and popular culture,
with interests in media engagement, everyday life, genres, production studies,
and cultures of viewing.
Akane Kanai PhD, Monash University) is Lecturer in the School of Media,

Film and Journalism at Monash University, Melbourne, Australia. Her
research interests include gender, race and affect; popular culture and digi-
tal identity cultures; and the transforming relations between femininity and
feminism.
Marina Krcmar (PhD, University of Wisconsin-Madison) is Professor and

Associate Chair at Wake Forest University. Her research examines how chil-
dren and adolescents use and are affected by media. Recent work has focused
on both traditional and social media, with an emphasis on moral reasoning
and well-being outcomes.
Rebecca Ann Lind (PhD, University of Minnesota) is Associate Professor

and Director of Undergraduate Studies in the department of communication
at the University of Illinois at Chicago. Her research interests include race,
gender, class and media; new media studies; media ethics; journalism; and
audiences.
Sharon Meraz (PhD, University of Texas at Austin) is Associate Professor

in the department of communication at the University of Illinois at Chicago.
Her research interests reside in the interplay of political communication, net-
worked journalism, and mass media theory.
Peta Mitchell (PhD, University of Queensland) is Associate Professor in

the Digital Media Research Centre at Queensland University of Technology,
Australia. Her research focuses on digital geographies, locative media, and
geoprivacy in everyday digital media use.
Brenda Moon (PhD, Australian National University) is a data scientist at

the Digital Media Research Centre at Queensland University of Technology,
Australia. Her research interest is using interdisciplinary approaches to apply
and develop digital methods for the study of public communication.
260 Contributors
Michael Potts (PhD, University of Canterbury) teaches communications and

sociology at the University of South Australia in Adelaide. His research inter-
ests include race and class in environmental literature and the culture and
literature of the twentieth century.
Erin L. Spottswood (PhD, Cornell University) is Assistant Professor in the

department of communication at Portland State University. Her research
focuses on how social media use can affect people’s interpersonal communi-
cation and relationships.
Lauren Taylor is a PhD student in the Department of Communication at the

University of California, Davis. Her research centers on media effects, specif-
ically with an interest in well-being and mental health outcomes.
Ellen Watts (PhD, Royal Holloway, University of London) is an LSE Fellow

in qualitative research methodology in the department of methodology at the
London School of Economics. Her research focuses on politics and popular
culture, particularly celebrity, representation, and power.
Yifan Zhao is a PhD student in the Department of Communication at the

University of California, Davis. His research centers on understanding how
the use of social media impacts emerging adults’ well-being.
Index
49ers  55, 78 Ahmed, Sara 50, 57–58, 59, 145–46

AIK. See Atheists in Kenya (AIK)
Alberoni, F.  34
algorithmic moderation  162
A algorithmic scandal 2–3, 166
#alllivesmatter 74, 80, 82, 100, 251
A21  133
Amazon, and religious influencers  127
Abidin, Crystal  14
Anderson, M.  198
abject beings  99–100
Andreassen, R.  143
Abriss, Erik  60
Ang, Ien  144
Accitelli, L. K.  202
Annabel, Eunice  37, 40
active use of media 196, 200–1, 203
antagonism, social media debates 2–3,
active broadcasting behavior  203
36–37, 40–41, 167–72
behavioral definition  201
anti-racist speech-acts  57, 58
positive outcomes  203
See also White anti-racism
affect 212, 214
appropriate constituencies  15
and emotional labor  213
appropriations 230, 248–50, 253
as a patterning 214, 221
Arnott, D. C.  197
theories on  212
asocial, humans as  231
affective practices  214–15
assemblage 212, 213–14, 229
of sociality and identity  140
and affect and emotions  214
and Utopia 222–24, 226
of geo-politics and sonic art
African Americans 50–51, 94, 113,
practices  226
115, 251
human-nonhuman assemblage
agency  231–32
213–14, 218–19
produsing as a matter of joint
of memories  218
agency  232
sound assemblage  217–18
age-related patterns, in social media
theories of  213–14
use  198
See also audience assemblage
agonism, social media debates  167–72
262 Index
Atheists in Kenya (AIK)  98, 99 B

attention economy 5–6, 35, 40–41, 45
attention events and rituals 5–6, 31–32, Banet-Weiser, Sarah  150–51
35–36, 40–41, 43, 44–45 Barker, Tim  217
attention-seeking  98–99 Barzilai-Nahon, K.  69
attractive/aversive attention  35 Baum, K.  196–97
audience assemblage 211, 212, Baumann, A.  196–97
213–14, 224–25 Beauchamp, Zack  61–62
affects and emotions  214 Beck, J. C.  35
conceptual foundation for  214 Benjamin, R.  118n4
glitchy palette  222–23 Berlant, Lauren 140, 143, 144–45,
and lines  214 146, 151–52
and patterning 214, 225–26 Bevan, J. L.  203–4
Utopia (See Utopia) Beyoncé 125–26
audiences 140, 211, 212 Lemonade album  78
audience information systems Biddle, Sam  49
214, 222 bigotry, crowdsourcing of  54–56
and celebrities  13–15 big social data 2–3, 159, 172, 173
celebrity-like forms  146–47 human intervention  160
as produsers  214 Bird, E.  41
texts and power 146, 151–52 Bird, S. E.  140–41
and their social positioning  144 Black Desert Online 236–37
and users, tensions and distinctions #BlackLivesMatter 67, 100, 251
between  143–44 across countries  68
audio-visual experience 220, 222, delegitimization of  80
223–24, 225–26 networked framing  68–71
Australia, political discussions in networked framing with
Twittersphere  165 hashtags  74–75
agonism and antagonism  167–72 calling attention to Black lives
Racial Discrimination Act, Section 18C lost  75–76
158, 166–67, 169, 173 of collective memory  76
#RoboDebt  166 of connecting diverse minority
authenticating appearance, influencer publics  77
wars  38, 39 of criticism  80
authenticity of current events significant to Black
authentic ambassador 12–13, lives  78
20–21, 25 as growing global movement  77
and feminism 140, 149–50 of police brutality  75
influence wars  31–32 of Trump campaign as trolling
legitimacy form 246, 249–50 #BlackLivesMatter publics  80
machines as authentic agents  240 of US presidential election
authentic voices  250 (2016)  78–79
avatars, Mattering framework networked gatekeeping 68, 72–74
233–34, 237 power law in  72–73
configuration permutations  236–37 power of networks to create meaning
continuity permutations  235 on  81–82
coordination permutations  234–35 tweets, sampling and analysis of  71–72
Index 263
and US presidential election and digital media  13–15

(2016)  67–68 faith-filled celebrities, as
Bless Box  134–35 influencers  131
Blogs, Wikipedia, Second Life and female celebrity texts  147–48
Beyond: From Production to honorific status of  14
Produsage 254 microcelebrities 14, 31,
legitimacy in  246 35–36, 213–14
#bluelivesmatter 74, 80, 82, 251 and politics 14, 250
bodily experiences 212, 213 and politics, relationship between  12
body image concerns  200 power, and user-audience
Boorstin, D. J. 35–36, 43–44, 249–50 networks  25–26
Bore, Inger-Lise 140, 150 as public text  147–48
borrowed interest 131–32, 253 public texts and everyday identity
Bortz, Rabbi Daniel, use of social media practice  152
129, 135–36 religious social celebrities  131
Botnik  229 texts and self-textualization  148–51
bots 7–8, 67–68, 70, 71–72, 73, 78, textual circulation of on social media
81–82, 126 144, 151
Bourdieu, P.  14–15 textualization and female
boyd, d. 142–43, 160 celebrity  147–48
branded content  122 celebrity capital 5, 12, 13, 15, 24
and the rise of social as strategy to move between fields 12,
influencers  123–25 13, 15, 25
branding 126, 149, 150–51 translating to political capital
Brown, Ashley  134 12, 15, 25, 250
Brown, Michael  67 Centrelink  166
Bruns, Axel 1–2, 69, 159, 214–15, 245, Chant, Annie 90, 97–98
246, 251, 253–54 Chicago School
Burgess, J. 36–37, 40–41 vision of Great Community  102
Butler, Judith  99–100 Clark, M. D. 68, 70–71
Buxmann, P.  201 Clarke, Adele  246
classical nerd scheme  106
class identity 7, 50–51
C class status  63
clickbait 39–40, 45–46
Caine, Christine  133 Clinton, Hillary  78–79
Callahan, B.  113 Coates, T.  1
Calo, R.  70 cognitive social comparison and social
capital, and competition  14–15 presentation  201
Carey, James  102 Coleman, Gabriella  107
Carpentier, N.  160 collective discourse  170–71
causality, and social media  203–4 collective fantasies  150
Cavell, Stanley  87–88 collective identity  144
celebrities collective meanings 140,
and audiences  13–15 142–43, 151–52
celebrity-like forms  146–47 collective memory, networked
and community  15–17 framing of  76
and desirable femininity  148
264 Index
collective participation in discourse  143 D

collective voice  97–98
Collins, M.  78 Dalai Lama, use of social media 128, 129
Columbia Journalism Review 62–63 Dallas 144
commentators  59–61 Dark Web  245
communication legitimacy  246
computer-mediated communication Davenport, T. H.  35
(CMC) channels 178–79, 183 Deadspin 55
of emotions, and legitimacy  253 Debord, G. 35–36, 43–44
human-machine relations  230 de-contextualization in the
as symbolic ritual  102 re-readings  152
community, and celebrity  15–17 Delanda, Manuel  213–14
competition, and capital  14–15 delegitimation 248–49, 250–51,
computational propaganda 70, 82, 249 252, 253
computational turn 157, 172 Deleuz, G.  213–14
computer-mediated communication democracy  88
(CMC) channels 178–79, 183 and discourse  167
configuration, in human-machine importance of speech in  87–88
association 233, 234–35, 236–37 as social cooperation  91–92
conflicts, produsers-involved  248 demographics  205
contents, of media  142 Dewey, John 88–89, 91–92, 102
continuity, in human-machine association Dhir, A.  198
233, 235 DiAngelo, Robin  56–57
controversy-seeking influencers 31–32, digital activism. See online activism
35, 41, 42, 43–46 digital culture 140, 142
coordination, in human-machine audience (See audiences)
association  233 branding in  150–51
corpus of tweets 157–58, 165, 170–71 disclosures  145
Couldry, Nick 88–89, 90–92, 93–94, 97, interactivity in  141
98, 99, 101, 141 and logics of visibility  146–47
counterfeit U.S. currency vendor’s textualization in  151
forum  245 digital distance  25–26
counter-normativity, and hating  42 digital identity 140, 146–47
covert marketing  122–23 digital intimacies 140, 143, 151–52
Crabtree, Michael  55 digital texts. See texts
Crawford, K.  160 discourse, and intimacy  145–46
cue-lean game formats 232–33, 236–37 discourse analysis  157
Cullors, Patrisse  67 challenges for  158–60
customer feedback  245 cyclical mixed-methods  161–65
cyclical mixed-methods social media practical applications for  165–72
discourse analysis  161 data collection  158
data capture  161 enlisting computational methods
generic analytics  161–62 in  172–73
practical applications for  165–72 and network analysis  170
social network analytics  163–64 platform-specific of
textual analysis  164–65 discourses  158–59
cyclical spectacles  43–44 discursive alliance  167–70
Index 265
discursive conditions  146 faith, as a tool in selling  127–28

discursive–material knot  160 faith-filled celebrities, as
distributed voice  98–99 influencers  131–32
diversity advocacy 107, 108–9, 110–11, fake news  67–68
113, 114–15, 116–17 false profits 133–34, 135–36, 253
DIY collectivities, and structural false social imaginary  224–25
change  117–18 family, impact on Influencers  33–34
Dobson, Amy 144–45, 146 Fanon, Franz  97
double consciousness 88–89, 97 Fearless Girl statue  124
downward social comparison  201–2 feedback loops  46
Driessens, O.  13 in hyperperception model  183
Droga5  124 in hyperpersonal model  178–79
Du Bois, W. E. B. 88–89, 97 female audiences
Dunbar, R. I.  186 celebrity and desirable femininity  148
Dungeons and Dragons Online 235 femininity/feminism 4, 90, 95, 99–100
audience-text approach  140
authenticity  140
E campaigns in Twitter 16 (See also Our
Shared Shelf (OSS))
echo chamber  54 feminine self-representation  140
elite class privacy  34 feminist audience 140, 143
emotional labor  213 in hacking and open source
Entman, R.  70 technologies 108, 110, 111–12,
Esposito, E.  160 115, 117
evangelicals/televangelists, as self-textualization  148
influencers  132–33 textualization  146
Evans, Rachel Held  130 Watson’s feminist politics (See
EVE Online 236–37 Watson, Emma)
exchange, legitimacy  248–49 and youth  146–47
#Ferguson  67
fictional texts  144
F FIKA. See Freethinkers Initiative
Facebook  195 Kenya (FIKA)
age-related pattern of use  198 Flanagan, Caitlin  61
motivations to use  197 Flinchum, J.  54
negative effects  203–4 FLOSS. See Free/Libre and open source
negative impacts  200 software (FLOSS)
and social comparison  202 framing  70–71
Facebook groups 87, 250 See also networked framing
Freethinkers Initiative Kenya (FIKA) Francis, Pope, use of social media
89, 102 128, 129
Pan-African Network (PAN) 90, 101 Free/Libre and open source software
voices  101 (FLOSS)  107
Women Without Religion (WWR) diversity advocates  110–11
90, 101 hackerspaces  252–53
face-to-face (FtF) interactions participation  107
178–79, 203–4 Freelon, D. 68, 70–71
266 Index
Free Software Foundation, gender sites of  107

disparity in 105–6, 109–12 voluntarism, open technology
Freethinkers Initiative Kenya (FIKA) communities  115–17
89, 102 as a worldview  105–6
distributed voice  98–99 Halliday, Aria  151
individual voice  96 Hardacker, C.  41
voice Harrison, K.  143
as an embodied process  94–95 hashtags  71, 72
as a form of reflexive agency  92–93 calling attention to Black lives
voice denial by voice-denying lost  75–76
rationalities  100 of collective memory  76
front-of-mind/back-of-mind of connecting diverse minority
attention  35 publics  77
functional agency  231–32 of global movements  77
Fyre 122–23 on police brutality  75
of US presidential election
(2016)  78–79
G Hathaway, C.  113
hating, among influencers 40–42, 43–44
Galatz, Chaviva 131, 135–36 counter-normativity  42
Gamson, W. A.  70 manufacturism  42
Garza, Alicia  67 non-news  42
gender disparity, in open source sensationalism  42–43
technologies and hacking 108, 109, temporality  43
110, 111–12, 115, 117 HeForShe conference 11, 15–16
generative justice  117–18 Hemmings, Clare  145–46
Geraghty, C.  34–35 Hendrix, Sazan  134–35
GIFs 38, 139, 144, 147, 148–49 heterarchical, bottom-up approach, to
Gillespie, T.  146 produsage  249
Goh, D.  53 highlight reels  202
Goldhaber, M. H.  35, 45 high-status privacy  34
Goldman, Andrew  59 Hill, Andrew  220
Gray, J.  200 Himelboim, I.  167–69
Green, J. 36–37, 40–41 Hochschild, Arlie 145–46, 213
Grenzfurthner, J. 105, 106 hooks, bell  87
Griffin, Kathy  60 Horowitz, Jake  53
Guattari, F.  213–14 Howard, P. N.  70
hug box  54
human-machine produsing 230, 231–33
H communication  230
hackerspaces  107 Mattering model for (See Mattering)
hacking communities 105, 117 potentials of  239
challenges of/to  113 Humphreys, L.  44
gender disparity in 106, 109–13 hyperperception model 3, 177–78,
growth potential in  105–6 179–80, 253
and I-methodology  113 anticipated and unanticipated
participation  107 consequences  189–90
Index 267
channel 180–81, 185 evangelicals/televangelists  132–33

feedback loop  183 faith-filled celebrities and
future directions  190–91 industrialists  131–32
meaningful conversations  189–90 false profits  133–34
observed receivers 180, religious innovators  129–30
182–83, 187–88 religious social celebrities  131
observed senders 180, traditional religious leaders  128–29
181–82, 186–87 rise of  123–25
observer  180 types of  125
and passive internet use (PIU)  190–91 value of  126
preliminary test of  183–84 influencer wars  36–37
channel  185 authenticating appearance  38
observed receivers  187–88 escalation  39–40
observed senders  186–87 productive disorder  39–40
hyperpersonal model  178–79 hating  40–44
web amnesia  44–45
status claims  37
I tell-all exposés  38–39
Ingold, Tim 211, 212–13, 214,
images 218–19, 224–25
-focused behaviors  198 inheritance  248–49
image effects  205–6 and legitimacy  253
and relational proximity  202–3 Instagram 31, 195
versus text, social media effects age-related pattern of use  198
198, 205–6 influencer wars  37
image-sharing  198 negative impacts  200
imagined audience  142–43 Internet Advertising Bureau (IAB)  122–23
I-methodology 6, 113–15, 117, 252–53 interpretive repertoire  220
individual voice  96 intimacies
industrialists, as influencers  131–32 de-traditionalization and
influencer marketing 122–23, 135 complexification in intimate
influencers 31–32, 122–23 relationships  143
attention events and rituals  35–36 and disclosures  145
commoditized privacy  32 and discourse  145–46
high status  34 effect of publicness of the circulation
low status  33 of  144–45
mid status  33, 34 as normative and regulatory relation
privacy for profit  34–35 143, 144–45
contemporary  249–50 intimate publics 140, 146
controversy-seeking 31–32, 42, 44
definition of  122–23
and followers  135–36 J
followers of  125–26
impact of family on  33–34 Jakes, T. D.  133
mega influencers  126 Jeffries, A.  52–53
micro-influencers  126 Jenkins, Henry  248
nano influencers  126 Jeong, Sarah  61–63
religious influencers  126–27 Joormann, J.  203–4
268 Index
K of produsage  245–46
propriety  253
Kabugi, Muchiri  92, 93 racialism  250–51
Kaepernick, Colin  78 symbolic economy of 246, 248–49
Kawalcyzk, C. M.  202 Lewin, K.  69
Kee, K.  53 Livingstone, S. 141–42, 151–52
Keller, Jessalynn  146 low-status privacy  33
Kelty, Chris  112–13 Lup, K.  202–3
Kenya  89
FIKA (See Freethinkers Initiative Kenya
(FIKA)) M
Khan, Abraham Iqbal  55
KhosraviNik, M.  160 Ma, L.  53
Kleemans, M.  200 machines  229–30
Kowal, Emma 50, 57, 58 advancement of abilities  239–40
Krasnova, H. 196–97, 201 and human assemblage
Krause, H. V.  196–97 213–14, 218–19
Krishnamoorthy, M.  113 see also human-machine produsing
Kruse, L.  54 MacIntyre, Alasdair  88–89
Maheshwari, S.  126
MailChimp  124
L Malik, A.  198
Manovich, L.  160
l8r h8r (later, hater) 31–32, 46 manufacturing controversies in digital
See also influencers media  54–56
Laclau, E. 165, 167 manufacturing outrage  52
Lady Gaga  125–26 manufacturism, and hating  42
and Intel  124 MapleStory 235
Latour, Bruno 230, 240 Marbles, Jenna  125–26
Lawrence, Jennifer 140, 147–49 marketing 121–23, 135
Layton, Danny 217, 218 branded content and social
Lee, C.  53 influencers  123
legitimacy covert marketing  122–23
academia  251–52 evangelicals/televangelists  132
authenticity  249–50 faith-filled celebrities and
authentic voices  250 industrialists  131
in Blogs, Wikipedia, Second Life false profits  133
and Beyond: From Production to influencer types  125
Produsage 246 native advertising  122
celebrity and politics  250 online and offline marketing
communication of emotions  253 components, integration of  124
definition of  246 religious influencers  126
digital research methods  252 religious innovators  129
and government 250 religious social celebrities  131
hacking  252–53 and traditional religious leaders  128
hashtags usage  251 market logics  115–16
human-machine interactions  252 Marshall, P. D.  13
and inheritance  253 Marshall, Solomon  99
Index 269
Martin, Trayvon  67 mixed-methods approach. See cyclical

Marwick, A. E. 21, 34–35, 142–43 mixed-methods social media
massively multiplayer online games discourse analysis
(MMOs)  234 Modigliani, A.  70
mass media 14, 67–69, 70, 81, 82, 141, Moores, S.  212–13
143–44, 197 Morrisey, Jack  60
Mattering 230, 232 Moss, Kirby  51, 63
avatars and players  233–34 Mouffe, C. 165, 167
configuration permutations  236–37 movement, theories of  212–13
continuity permutations  235 Mukbang internet videos  213–14
coordination permutations  234–35 Munden, Marc  217
forms and implications  237–38 Muslim Travel Girl 131
legitimacy  252
produsing as a matter of joint
agency  232–33 N
configuration  233
continuity  233 Nakamura, Lisa  151
coordination  233 nano influencers 126, 129
as social and agentic  230–32 Nation, The 55
subjective experiences of  239–40 National Urban League  251
McEwan, B.  179 native advertising  122
Mcllwain, C. D. 68, 70–71 negative attention rituals 5–6, 31–32,
media contents 1–2, 3–4, 100, 102, 142, 35–36, 40–41, 43, 44–45
197, 206, 229 Neiborg, David  247
media framing  70–71 neoliberalism  88
media gatekeeping  68–70 network analysis, and discursive
Meeker, Mary  125 analysis  170
mega influencers  126 networked framing  70–71
Meier, E. P.  200 of #BlackLivesMatter as a growing
memories, assemblage of  218 global movement  77
Mendes, Kaity  146 of calling attention to Black lives
Mendick, H.  24 lost  75–76
Meraz, Sharon  69, 70 of collective memory  76
Meyer, Joyce  133 of connecting diverse minority
Mic 52–53, 54 publics  77
microcasts  142–43 of criticism  80
microcelebrities 14, 31, 35–36, 213–14 of current events significant to Black
influencers  31 lives  78
micro-influencers 126, 129 of police brutality  75
middle-class White identity 51, of Trump campaign as trolling
52, 56, 63 #BlackLivesMatter publics  80
mid-status privacy  33 of US presidential election
Miller, Judith  52 (2016)  78–79
Minaj, Nicki  151 networked gatekeeping 68–70, 72–74
minoritized populations  151 network formation  167–69
minory publics (diverse), networked news, as cultural product  59–60
framing of  77 Ng’ang’a, Daniel  92–93
Nichols, Alex  54
270 Index
Nieminen, M.  198 as connected representative  18–19

Nikolova, Elena  131 as ordinary member of OSS  19–20
Nkrumah, Kwame  100 user-audiences evaluation  22–24
nomination, discursive strategy  171–72 over-attributions  179
nonhumans  231 Owen, David  88
See also machines
non-news, and hating  42
non-representational theory  212 P
nonverbal cues, in hyperpersonal
model  178–79 Pai, P.  197
normative femininity  110–11 PAN. See Pan-African Network (PAN)
Norris, D.  54 Pan-African Network (PAN) 90, 101
voice as a form of reflexive agency  92
voice denial by voice-denying
O rationalities  100–1
Papacharissi, Zizi 2, 34, 69, 70
observed receivers, in hyperperception parasocial interactions 14, 231, 239
model 180, 182–83 Park, N.  53
observed senders, in hyperperception Park, Robert  102
model 180, 181–82 participatory web  159
observer, in hyperperception model  183 passive internet use (PIU)  190–91
Okullu, Henry  98–99 passive use of media 198–99, 200–1, 203
Olufadi, Y.  200–1 behavioral definition  201
Omare, Annette  94 peer production  105–6
online activism  67 Perrin, A.  198
online and offline marketing components, Petersen, M.  143
integration of  124 PewDiePie  125–26
online media, functions of  54 Philips, Nathan  60, 61
open technologies  107 Phillips, W.  40–41
communities (See also hacking photo-filtering  200
communities) pink technology  110–11
and politics  115–17 Pinterest  150
voluntarism 114, 115–17 Pittman, M. 199–200
diversity advocacy 107, 108–9, 110– platform cultures  146
11, 113, 114–15, 116–17 platforms 157, 213–14
gender disparity in  109 algorithmic design of  162
gender in  111 big data from  160
women of color in  113–14 demographic differences  198
OSS. See Our Shared Shelf (OSS) discourse analysis  157–58
Osteen, Joel  133 and discourse analysis  158–59
Oudshoorn, N.  115 feeling rules of  150
#OurSharedShelf  21 impact on societal
Our Shared Shelf (OSS) 5, 11–13, 250 developments  159–60
affordances of  15–17 motivation to use  198
and Watson  17–18 specific features and affordances
Watson's representative claims  18 of  196
as authentic ambassador  20–21 text- and image-based  199
Index 271
uses and gratification of 197 public shaming 49–50, 51–52

(See also uses and gratifications purchase 125, 127–28,
framework) 195–96, 248–49
player-avatar associations. See avatars,
Mattering framework
Plummer, Ken  143 Q
polarized network  169–70
police brutality, networked framing of  75 Quan-Haase, A.  197
PolicyMic 52–53, 54, 55 quotidian movements of audiences and
political capital 12, users  212–13
14–15, 19, 25, 250
political power, and celebrities
R
relationship with audiences  14
Polletta, Francesca  114 Racial Discrimination Act (Australia)
popular cultures  229 Section 18C 166–67, 169, 173
Portman, Emily  94–95 racial etiquette 50, 51, 56, 250–51
positive self-presentation  183 racial identity 50, 57, 250–51
Pounders, K.  202 Rainie, L.  167–69
power Rambukkana, Nathan  145–46
audiences and texts  151 Raun, T.  143
law of  72–73 reaction-GIF texts 139, 144, 145–46,
relations, in technological 147, 148–49
communities  117 receivers
@preachersnsneakers  130–31 in hyperperception model
predication, discursive strategy  171–72 180, 182–83
privacy, as commodity  32 in hyperpersonal model  178–79
high status  34 re-contextualization in the
low status  33 re-readings  152
mid status  33 redressive action  40
for profit  34–35 Reese, S.  70
processing of media content  206 Reich, B. 199–200
productive disorder, influencer wars reintegration process  40
as  39–40 Reisigl, M.  171–72
hating  40–44 relational proximity  196
web amnesia  44–45 and social media  202–3
produsage 1–2, 69, 229, 237, 245, 247 religion
programmatic ads  123–24 and marketing 121–22, 135
propaganda theory  249–50 as a product  121–22
Propel Women  133 and social media 128, 135
propriety, legitimacy form  246 religious influencers  126–27
prosumer  214–15 evangelicals/televangelists  132–33
Protection of Harassment Act of 2014, faith-filled celebrities and
Singapore  43 industrialists  131–32
pseudo-events 5–6, 35–36, false profits  133–34
43–44, 249–50 religious innovators  129–30
public forum  59 religious social celebrities  131
publics as affective spaces  145–46 traditional religious leaders  128–29
272 Index
religious innovators, as influencers  129–30 Shoemaker, P.  69

religious social celebrities, as Slate 49–50
influencers  131 Slaton, Amy  117
reporters  59 Smith, Dorothy 97, 169
retouching 199, 200 Smith, M. A.  167–69
retweet networks  169–70 SMRT Feedback Ltd  43
Ringrose, Jessica  146 Snapchat  195
#RoboDebt 2–3, 158, 166, 167–69, 173 age-related pattern of use  198
Rojek, C.  14 negative impacts  200
Ronson, Jon 49–50, 52 social
Run the Race, promotion of  132 anthropocentric view of  231
Russia mattering as  230
Internet Research Agency  251 social anxiety 190–91, 203–4
social attractiveness  178–79
social comparison 201–2, 204, 206
S downward comparison  201–2
upward comparison  201–2
Sacco, Justine 49–50, 51 social distance 13, 14, 15
Salam, Reihan  63 social dramas 5–6, 36, 44
Salon 63–64 social imaginary 221–22, 224–25
Sanders, Bernie  78–79 social influencers. See influencers
Saward, M.  15 social media  54
Schneider, F. A. 105, 106 audiences direct interaction with
schoolboy and Native American protestor, celebrities  14
confrontation between  60–62 channels 179 (See also
Schumer, Amy 140, 147–48, 150 hyperperception model)
Seattle Seahawks  55 cognitive processing  196–97
selective self-presentation 179, 202 discourses  160
self-expression  151 horizontal scale  160
self-presentation 13, 18–19, 24, 142–43, vertical scale  160
179, 183, 202 effects 195–96, 198
self-representation 38, 140, and feminist activism  16–17
142, 147, 149 micro-celebrities  14
self-surveillance  44 news stories  53–54
self-textualization  148–51 presentational culture  35
senders and relational proximity  202–3
in hyperperception model and religions 128 (See also religious
180, 181–82 influencers)
in hyperpersonal model  178–79 religious innovators  129
sensationalism, and hating  42–43 and self-esteem  196–97
sex content, influence over audience  33 social media platform. See platforms
shaming practices  45 social media use
shared imaginary  145–46 behaviors 196–97, 206
Shaw, A. M.  203–4 active and passive 200, 201, 203–6
Sherman, Richard  55–56 image-focused behaviors  198
Sherman, Tracy  91 self-representation  202
Shmuley, Rabbi  132–33 cognition 196–97, 201
Shneiderman, B.  167–69 active and passive 200–2, 204, 206
Index 273
motivations of 197–98, 199 female celebrity as public text  150

patterns  196 fictional texts  144
prediction of  197–98 versus images, social media effects  198
and representation  14–15 production of self as  146–47
social media use, and well-being  195–97 textual circulation, of celebrity on social
active and passive use  200–1 media 144, 151
self-presentation  202 textualization 146–47, 152
social comparison  201–2 celebrity texts and
causality direction  203–4 self-textualization  148–51
images versus text 198–200 in digital culture  151
model, proposal of  204–6 of the self  152
relational proximity  202–3 Thompson, M.  223–24
uses and gratifications research Thrift, Nigel  212
197–98, 252 Thumim, Nancy  142
social software 1, 2, 3, 6–7, 8 Tiidenberg, Katrin  146
Spectator, The 144 Timpano, K. R.  203–4
speech, importance in democracy  87–88 Tometi, Opal  67
Stacey, Jackie  148 traditional religious leaders, as
State of Black America 251 influencers  128–29
status claims, influencer wars  37 Tran, T. B.  203–4
status enhancement  53 troll  80
Steers, M. N.  202 definition of  41
stories, and voice  88 Trump, Donald  79
Stowers, K.  202 networked framing of his campaign
Streeter, Thomas  114 as trolling #BlackLivesMatter
subjects  99–100 publics  80
Sullivan, Andrew  61–62 Tumblr
Sullivan, Shannon  50 reaction-GIFs posts 139, 144, 145–
Swisher, Kara  60 46, 147, 148–49
symbolic economy of legitimacy  246 self-representation of young women
on  147
Turner, G. 14, 34–36
T Turner, V.  36, 39
Twitter  195
“take a knee “movement  78 influence wars  37
Tapia de Veer, Cristobal  217–18 network formation  167–69
Taylor, Charles 221, 224–25 openness of  82
Tebow, Tim  132 religion-related uses of  130
technoscientific innovation  118n4 See also Australia, political
tell-all exposés, influencer wars  38–39 discussions in Twittersphere;
temporality, and hating  43 #BlackLivesMatter; #RoboDebt
text effects  205–6
text-reader metaphor 140, 142, 146
collective participation  143 U
texts 141, 151–52
audiences and power  151 UN Women
celebrity as public text  147–48 HeForShe campaign 11, 15–16
circulation of  144 upward social comparison  201–2
274 Index
user-audiences voice 87, 101

evaluation of Watson’s representative authentic voices  250
claims  22–24 collective voice  97–98
networks  12 concept of  88
proximity  13–15 definition of  87–89
user gratifications 197, 204–5 denial by voice-denying rationalities
users  140 88–89, 99–101
and audiences, tensions and Freethinkers Initiative Kenya  100
distinctions between  143–44 Pan-African Network  100–1
uses and gratifications framework 197– Women Without Religion  100
98, 204–5, 252 distributed voice  98–99
US presidential election (2016), as embodied process 88–89, 93–95
networked framing of  78–79 Freethinkers Initiative Kenya  94–95
Utopia 211–12, 215, 250 individual voice  96
assemblage of geo-politics and sonic art and material form  88–89
practices  225 as reflexive agency form 88–89, 91–93
audiences Freethinkers Initiative Kenya  92–93
assemblage of  219–22 Pan-African Network  92
political and cultural Women Without Religion  92
engagement  222–24 requires a material form  96
audio-visual experience  220 Freethinkers Initiative Kenya
audio-visuals  223–24 96, 98–99
false social imaginary  224–25 Women Without Religion  97–98
geo-politics and sonic art as socially grounded 88–89, 90–91
practices  223–24 Women Without Religion  91
glitchy palette  219 and stories  88
reflections  224 voluntarism, open technology
remix project  218 communities  115–17
research process  216 voluntary/captive attention  35
social imaginary  221–22 Vos, T.  69
soundtrack  221 Vox 61–62
soundtrack creation  217–18
Utz, S.  200
W
V Waldman, Katy  56
Walther, J. B. 178–80, 183
Valenzuela, S.  53 Warner, Michael 34–35, 102, 144
Van Dijck, José 140–41, 247 Warren, Rick  133
Van Volk, Peter  95, 96 Washington Post, The 60–61
Vice 52, 53 Watson, Emma 11, 250
videogame players, Mattering and Books on The Underground
framework  233–34 project  21
configuration permutations  236–37 celebrity capital/status 12, 13,
continuity permutations  235 17, 20, 23
coordination permutations  234–35 and media platforms, use in Our
violence  151 Shared Shelf representation
legitimacy form  246 claim  25
Index 275
Instagram  21–22 Women Without Religion (WWR)

interview with Satrapi  19 90, 101
and Our Shared Shelf (OSS) 11 collective voice  97–98
(See also Our Shared Shelf (OSS)) voice
participation in feminist campaigns  21 as a form of reflexive agency  92
political capital  19 as socially grounded  91
social distancing 21–22, 24, 25 voice denial by voice-denying
social media following of 14, rationalities  100
15, 20–21 Woods, Salome  94
social media usage  21, 25 Woolley, S.  70
UN speech 11, 15–16 World of Warcraft 238
user-audience networks  12–13 WWR. See Women Without
Weaving the Dark Web 248 Religion (WWR)
web amnesia  44–45
Weismann, Jordan  49–50
well-being  195–96 X
and social media (See social media use,
and well-being) Xiaxue  37, 43
Wenninger, H.  201
Wetherell, Margaret 139, 145–46, 214,
Y
221, 224–25
White, Paula  133 Yang, C.  203
White anti-racism  49–54 Yizhak, Leopold  100
bigotry  54 Young, A. L.  197
White identity, policing of  56 young women  146–47
White identity  50 YouTube
middle-class 51, 52, 56, 63 flame wars 36–37, 40–41
policing of  56–63
Wickham, R. E.  202
Widjaja, T.  201 Z
Williams, Raymond  246
Wodak, R.  171–72 Zanolla, D.  179
women Zimmerman, George  67
bifurcated consciousness of  97
See also femininity/feminism
General Editor: Steve Jones
Digital Formations is the best source for critical, well-written books about digital
technologies and modern life. Books in the series break new ground by emphasizing
multiple methodological and theoretical approaches to deeply probe the formation
and reformation of lived experience as it is refracted through digital interaction. Each
volume in Digital Formations pushes forward our understanding of the intersections,
and corresponding implications, between digital technologies and everyday life. The
series examines broad issues in realms such as digital culture, electronic commerce,
law, politics and governance, gender, the Internet, race, art, health and medicine, and
education. The series emphasizes critical studies in the context of emergent and
existing digital technologies.
Other titles include:

Felicia Wu Song Edited by Judith Burnett, Peter
Virtual Communities: Bowling Alone, Senker & Kathy Walker
Online Together The Myths of Technology:
Edited by Sharon Kleinman Innovation and Inequality
The Culture of Efficiency: Technology in Edited by Knut Lundby
Everyday Life Digital Storytelling, Mediatized
Edward Lee Lamoureux, Steven L. Baron, & Stories: Self-representations in
Claire Stewart New Media
Intellectual Property Law and Interactive Theresa M. Senft
Media: Free for a Fee Camgirls: Celebrity and
Edited by Adrienne Russell & Nabil Echchaibi Community in the Age of Social
International Blogging: Identity, Politics Networks
and Networked Publics Edited by Chris Paterson & David
Edited by Don Heider Domingo
Living Virtually: Researching New Worlds Making Online News: The
Ethnography of New Media
Production
To order other books in this series please contact our Customer Service Department:
peterlang@presswarehouse.com (within the U.S.)

order@peterlang.com (outside the U.S.)
To find out more about the series or browse a full list of titles, please visit our website:
WWW.PETERLANG.COM

Produsing Theory in A Digital World 3.0 The Intersection of Audiences and Production in Contemporary Theory - Volume 3 (Rebecca Ann Lind (Editor) )

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

Produsing Theory in A Digital World 3.0 The Intersection of Audiences and Production in Contemporary Theory - Volume 3 (Rebecca Ann Lind (Editor) )

Uploaded by

Copyright:

Available Formats

REBECCA ANN LIND, editor

The Intersection of Audiences and

“A collection of sparkling ideas from many of the field’s best thinkers,

“Through a series of thought-provoking essays bound to appeal to

Rebecca Ann Lind (Ph.D., University of Minnesota) is Associate Profes-

The Intersection of Audiences and

Production in Contemporary Theor

The Digital Formations series is part of the Peter Lang

The Intersection of Audiences

Edited by Rebecca Ann Lind

Produsing theory in a digital world: the intersection of audiences

Includes bibliographical references and index.

1. Mass media—Technological innovations. 2. Mass media and technology.

P96.T42P76 302.23—dc23 2012014097

ISBN 978-1-4331-1520-2 (Volume 1 hardcover)

ISBN 978-1-4331-2729-8 (Volume 2 hardcover)

ISBN 978-1-4331-5339-6 (Volume 3 hardcover)

© 2020 Peter Lang Publishing, Inc., New York

9. Audiences, Affects, Attachments: Theorizing Textual Approaches to

produsage, briefly defined as “the collaborative and continuous building and

discourses—the authors show how their cyclical mixed-methods approach can

media content. The machines or assemblages with which Banks is concerned

wars can be understood as social dramas, pseudo-events, a form of productive

others attempted to delegitimate or even troll #BlackLivesMatter. By tracking

Coates, T. (2005, January 5). An addendum to a definition of social soft-

As part of my work with UN Women, I have started reading as many books

In September 2014, Hollywood actor Emma Watson stood in front of the UN

the milestone of 100,000 members within a month, Watson (2016d) thanked

Celebrity and Digital Media: Representing Proximate

The relationship between celebrities and audiences has, of course, evolved

Community and Celebrity: The Affordances

community-building, because the structure and slower pace of threads enable

22 participants—recruited through a message Watts posted on the forum—

1. Why did you want to join Our Shared Shelf?

Watson as Connected Representative

Watson as Ordinary Member of OSS

Crucially, this positioned Watson as fellow learner rather than all-know-

Watson as Authentic Ambassador

representative she foregrounded elite connections, while as ordinary member

How did User-Audiences Evaluate Watson’s Representative

Table 2.1. Perceptions of Watson as a Representative

Watson is perceived as Number of participants

Participants often discussed Watson by comparing her to “other celebri-

Digital Distance: Celebrity Power and User-Audience Networks

Indeed, we argue Watson’s balance of proximity and distance was a fun-

emmawatson. (2016a). Who has their book? #OurSharedShelf #BookClub

Marshall, P. D. (2014). Celebrity and power: Fame in contemporary culture.

Since their earliest commercial endeavors on blogs such as Blogspot or

to chase drama and controversy, garner negative attention for themselves or

Commoditized Privacy: A Lifecycle

congruence of the constructed narratives (Daniel & Knudsen, 1995). For

the other to bait followersʼ attention. As a form of “lifestreaming” (Marwick,

Attention Events and Rituals

Influencer Wars: Three Case Studies

of disorder, attempting to produce side commentaries, personal editorials, or

Haters repudiate controversy-seeking Influencers for being counter-nor-

Although Influencers frequently headline newspapers and magazines, haters

The third type of hating discourse focuses on the manufactured nature of

Haters also decry the sensationalist nature of Influencersʼ “antics,” citing

as soon as one Influencer (mis)interprets anotherʼs “vague” Tweet as a smear

Despite their denouncement of Influencersʼ controversy-seeking practices

(1961) pseudo-event, in which the generated news staged by controver-

Unlike scholarly discussions that describe the infrastructure and technology of

1. Why did you want to join Our Shared Shelf?

Watson as Ordinary Member of OSS

Calling Attention to Black Lives Lost

Networked Framing of Current Events Significant to Black Lives

A Brief Description of the Facebook Groups

Freethinkers Initiative Kenya (FIKA)

Women Without Religion (WWR)1

Voice as a Form of Reflexive Agency

Warner, M. (2002). Publics and counterpublics. New York, NY: Zone Books.