COLLECTIVE IDENTITY FORMATION AND

COMMERCIAL PLATFORM LOGICS IN SOCIAL

ACTIVISM: REPRESENTATION OF WOMEN AND
BLACK FEMINIST ACTIVISM ON INSTAGRAM UNDER
#BLACKLIVESMATTER

Ellimaija Tanskanen

Communication for Development

One-year master
15 Credits
Spring 2021
Supervisor: Tobias Denskus
16.8.2021
Abstract
Due to the participatory nature of social media platforms, users contribute to the narratives
built around online action for social change and shape the discourse on societal topics
through their participation. At the same time as social media has become a space for
societal activism and participation facilitating connective action of individuals, social media
platforms are ultimately, for most, owned by private companies. This makes them products
of the attention economy, where the attention of consumers has been quantified and
commodified and where different players compete for such attention. The current research
presents an analysis of content related to online advocacy to inform on the effects of a social
media platform on social change and the use of a platform by citizens. More specifically,
the research focuses on collective identity building through visual self-representation and
how the commercial structures of the platform and the participation of users affect the
representation of women in the context of the Black Lives Matter movement on Instagram.
Methodologically the research was performed through a quali-quantitative exploration of
publications associated to the hashtag #BlackLivesMatter, using cultural analytics and
content analysis. The research concludes that while the complexity of technological and
human variables in online societal participation makes the research on representations of
women challenging because of the various actors and forces at play affecting it directly or
indirectly, the hashtag #BlackLivesMatter is largely used for collective identity building that
can contribute to empowering marginalized groups on social media. This type of finding
nevertheless emphasizes the memetic characteristic of the hashtag rather than a tool for
direct social activism.

1
Table of Contents

Abstract ............................................................................................................................ 1
Introduction ..................................................................................................................... 3
Literature Review ............................................................................................................. 4
The Origin of Platforms, Platform Capitalism, and the Effects of Underlying Platform
Mechanisms on Social Media Use .......................................................................................... 4
Social Media Activism – A Disputed Tool for Social Change in the 21st Century ............. 6
Collective Identity in Social Media Activism ........................................................................ 7
Black Activism and Racial Justice in Online Social Change Processes .............................. 8
Intersectionality and Black (Cyber)feminism in Social Change ....................................... 10
Theoretical Framework – A Collective Identity Building Approach to the
Representation of Women in the Black Lives Matter Movement ..................................... 13
Methodology ................................................................................................................... 14
Data Collection ....................................................................................................................... 14
Data Analysis.......................................................................................................................... 15
Framework of Analysis ..................................................................................................................... 15
Cultural Analytics.............................................................................................................................. 16
Content Analysis ............................................................................................................................... 17
Results ............................................................................................................................ 18
Cultural Analytics.................................................................................................................. 18
Complete Dataset .............................................................................................................................. 18
Images of Women ............................................................................................................................. 19
Content Analysis .................................................................................................................... 20
Discussion ...................................................................................................................... 21
Instagram – The Challenges of a Commercially Based Online Environment for Social
Activism and Collective Identity Formation ....................................................................... 21
Limitations ............................................................................................................................. 29
Conclusion ..................................................................................................................... 31
References ...................................................................................................................... 34
Appendices ..................................................................................................................... 38
Appendix 1: Content Analysis Code Book .......................................................................... 38

2
 Introduction
Social media have become a part of everyday life for many in the 21 st century. In addition
to being a tool for communication, they can also be seen as a tool for societal action using
an interconnected network of users independent of space or time. While there are many
examples of successes related to online activism in generating change, social media activism
also presents a number of challenges, such as the role and power distribution between
different actors involved, or the real effects of social media action for social change. Much
of these challenges are ultimately based on how the platforms are built and why, and how
users adopt the platform features or affordances. Based on this, platform companies behind
such platforms yield a lot of power in shaping social media use. The decisions made by
platform companies in turn are strongly based on the commercial value of the platform.
Power dynamics and the impact of social media activism can be studied through the analysis
of representation, as representations contribute to shaping the narrative around a topic. In
societal activism on social media platforms, representations shape which discourses and
topics are brought forward and how. This research uses the case of the Black Lives Matter
(BLM) movement on Instagram to analyse discourses brought forward by users around the
topic. The research focuses on representations of women as one group of users with specific
topics of interest and needs in the BLM movement and activism against racial oppression
in general, to determine how the platform affordances and the underlying commercial
logics as well as users’ behaviour on the platform affect the narrative and collective identity
built around women on anti-racist social media activism. The Black Lives Matter movement
is a particularly timely topic of research, with close to 28 million publications on Instagram
under the hashtag #BlackLivesMatter at the time of the research. At the same time, it is an
example of a social movement related to a complex and multidimensional societal issue,
namely racial oppression, which affects different groups of people and in different ways,
involving many different types of stakeholders. Women, and specifically Black women and
women of colour, are minority groups having always had to fight for their rights and face
injustice on a variety of fronts, including but not limited to race, gender, and class. To
ensure that social movements reach their advocacy goals, it is therefore particularly
important to study how these groups in more vulnerable positions are represented in the
scope of the social movement and why. At the same time, this type of research can give
indications on the broader societal impact of social media content and their role in societal
activism.

3
I will first present a review of existing literature on social media activism, collective identity
construction, platform capitalism, and black feminist theories used in online social
movement research and tie this literature together into my theoretical framework. In the
next chapter, I will present my chosen methodology for the analysis, including cultural
analytics and content analysis, and a basis of discourse analysis. I will then present the results
of my analysis in a descriptive manner in the “Results” chapter. After this, I answer my
research question and reflect on the meaning of my results, tying them back to the
theoretical framework and my personal experience conducting the research. Finally, I finish
the research on a conclusion of the results and possibilities for future research.

Literature Review
In the following section, I will present the existing literature on which the current research
is based, starting from the background of social media platforms and the functioning of
platform companies. I will then elaborate on the role of social media in social change
processes and social movements and present the background and specificities of the Black
Lives Matter movement used as case study in this project. As a third element, I draw on
literature on intersectionality theory and Black cyberfeminism, which are pillars in modern
feminist social research and bring forward the importance of analysing representations of
women in a contemporary anti-racist social movement online. Finally, I present a
framework for the analysis based on the literature review to answer the following research
question: How is Instagram used to represent women as a part of the Black Lives Matter
movement?

The Origin of Platforms, Platform Capitalism, and the Effects of Underlying

Platform Mechanisms on Social Media Use

The last decade has seen the unbelievable growth and adoption of new technologies at a
pace never seen before in history. Since the popularization of the Internet in the late 20 th
century and the development of what is often referred to as “Web 1.0.”, or the static
structure of the Internet in which content creators share content to large audiences online,
the 21st century has pushed technological development a step further. In the beginning of
the century, new technologies started allowing an increasing amount of participation for all
Internet users. This second form of the Internet, characterized by user-produced content
and its participatory nature is, unsurprisingly, referred to as Web 2.0.

4
Comparably to decades earlier when techno-optimistic forerunners had already perceived
the potential of the Internet as it was popularized, in the first years of Web 2.0., the new
participatory nature of the Internet and the use of multimedia content brought hope for a
more democratic Internet controlled and moulded by ordinary citizens, as well as the fall
of the monopoly of traditional mass media companies (Srnicek, 2017; Zulli, 2020). These
hopes were nevertheless cut short with the rapid rise of what have become key names in
digital technologies – Google, Facebook, Twitter, and other platform companies started
building real empires on networked participatory platforms. These companies were able to
do so thanks to the collection, analysis and use of huge amounts of data, which has become
a key currency in the platform economy. This in turn has resulted in the monopolistic
power of a small number of companies having the best access to and usage of data (Srnicek,
2017). In the process, the limited attention of users in conflict with the now almost
unlimited amount of content has resulted in a new form of economic system built on
attention, as a variety of players have quantified, commodified, and started competing over
the attention of consumers (boyd, 2010). These players include advertisers, private
companies, other users, the media, and even organizations such as governments or third-
sector players alike.
A key for a platform company to gain such a power position in society is in the fact that at
its simplest, a platform makes the interaction of two actors possible. In this way, the
company stands in a broker position between these actors while at the same time this
interaction allows the company to record and use the data resulting from it (Srnicek, 2017).
Platform companies therefore play a crucial role in determining the use of the platforms by
consumers, and such decisions are for a major part driven by the commercial value of
platform affordances. These commercial logics are particularly important, as they play a
crucial role in the platform economy: to keep growing and maintain their competitive
position, platform companies must ensure that current users stay, and that an increasing
number of new users join the platform by developing constantly new tactics (Srnicek, 2017),
including a better use of the platform affordances and developing algorithms, to grow the
commercial value of the platform. To conclude, the conceptualization of social media use
in this research is based on three pillars: the technological structure of the platform, the way
users interact on it, as well as the commercial value of the platform.

5
Social Media Activism – A Disputed Tool for Social Change in the 21st Century

While the online sphere is for a large part built around commercial platform companies,
this does not take away the participatory nature of such platforms and their wide and varied
use by citizens for social, entertainment or political reasons for instance. On the opposite,
platforms are dependent on such interaction, and for this reason also aim to present
themselves as empty spaces where interaction can be freely built between users to make the
platform more appealing to them, while in reality the rules and interaction is set by a politics
controlled by the company (Srnicek, 2017).
Because of this, contrary to frequent PR statements by platform representatives when
accused of action (or the lack thereof) in societal events built on social media interaction,
platforms do not bear a neutral position in social change and activism online. Social media
activism, defined by Brown and colleagues (2017) as “the use of social network technology
to organize and coordinate real-world action”, is an important tool in today’s social
movements and organization of engagement and events related to political activism. Social
media activism can be seen as a part of “mediatization”, a process entangled in societal
trends such as globalization and individualization, in which media is becoming an
increasingly important element in individuals’ lives, but also in organizations’ and
institutions’ activities (Tufte, 2017). Nevertheless, private organizations and companies
remain main drivers of the use of media in society. While private companies are also drivers
of social change and many cases demonstrate successful endeavours, their activities are
ultimately based on a market-driven economic growth model, with a strong aspect of
Western-centrism and consumerism (Tufte, 2017). This is on many aspects in line with
those of platform companies, but often criticized regarding their motives and effects on
society.
On a citizen perspective though, social media has facilitated the participation of people
across time and space, as well as given the possibility for marginalized groups, such as youth
for instance, to gain a stronger voice in social change processes (Tufte, 2017). Partly for this
reason, they are increasingly used as additional tools for social change and movements next
to more traditional forms of civic activism (Haunss, 2015). Simultaneously, social media
activism’s real effects on society are often contested, and online activism has been criticized
as “clicktivism” or “slacktivism” with little real impact (Christensen, 2011; George &
Leidner, 2019). While the effect and efficiency of different forms of online activism may
vary, social media have indisputably had a crucial role in societal participation and

6
processes of social change, both as a tool to reinforce traditional forms of activism, and as
a mean of its own. Platforms are “both sociotechnical constructions and tools for building
larger social and political structures” (Avram et al., 2019) and undeniably influence society.
Whether that effect is positive or negative is disputed, and platforms’ functioning must be
researched further from a societal perspective to avoid pitfalls and turn platforms into tools
for positive change.

Collective Identity in Social Media Activism

Social activism relies on the power of masses to mobilize to create change. Compared to
traditional, organization-led activism, group mobilization in the digital space is based rather
on the networked structure of the platform. Bennett and Segerberg (2012) presented for
this an updated concept for collective action for online social movements – the concept of
“connective” action describes the characteristic of online activism based on self-organized
action, which derives rather from personal interests of individuals in a specific cause rather
than a common core ideology or group identity between members. In this way, group
formation for online activism tends to be rather circumstantial and built around a specific
topic or cause, which also both fits in the ever-changing and quickly evolving networked
structure of a platform, and emphasizes the fluidity of social movements online.
Collective identity nevertheless remains an important concept for social movement
research even in the study of more connectivity-based action, but classic definitions of
collective identity based on the sense of belonging to a specific group need to be revised to
fit in the context of online activism. Kavada (2015) perceives collective identity as a dynamic
process that is constructed through conversation, which blurs on social media the limits of
the ingroup and outgroup. Collective identity construction can be seen to include two
elements: discursive processes, which are based on narratives and meaning making to create
shared understandings of a situation, and enacted processes, which activate the relationships
between members and include an element of emotional investments in the identity
construction (Khazraee & Novak, 2018). Visual representations play an important role in
collective identity contruction (eg. Gerbaudo, 2015), and particularly so in the enacted
processes of collective identity building (Khazaree & Novak, 2018).
Similarly, Einwohner and Rochford (2019) consider activism-related social media
publications to be a binding factor in collective identity that facilitates social movements.
This is done through the creation of a group experience through the communication of
individual users, which is facilitated by platforms through their networked structure and

7
user participation. Particularly on a platform based on imagery with a reputation of being
“positive” and “entertaining” such as Instagram, political action can be considered a
performance that aims to display identity and identify oneself to a certain group (Einwhoner
& Rochford, 2019). The authors further state that, based on Goffman’s (1959) performance
theory, the participatory aspect of the platform and its appeal to young women “may
encourage gendered political performances” (Einwohner & Rochford, 2019). Similarly,
Wilkins and colleagues (2019) present that the public and argumentative nature of social
media platforms allows “users to attempt to shape and contest social identities in the face
of alternative positions” and works as a space of debate over the importance and essence
of a social movement while mobilizing support. The role of the platform in this equation is
thus not meaningless, and it is not by coincidence that social media foster social movements
and activism.
Social media platforms’ varying affordances affect collective identity-building processes as
per Wellman’s (2001) concept of social affordances technological changes affect the
formation and possibilities for social relations and social structure. Instagram is
characterised particularly by its technological structure based on visuals, which makes it a
highly relevant platform for the study of collective identity formation. The affordances of
the platform and their use has resulted in that the visual composition of an image is more
important to engagement than the corresponding caption or hashtag (Zulli, 2018). A s a
platform focused on community-building and identity, associated with the visuality of the
platform, representations and framing of publications matter strongly. The analysis of
representations and framing of content on Instagram therefore contributes to a better
understanding of the role of Instagram in social change processes when analysing content
specifically centred on social media activism. The finding in previous research that there is
a growing focus on visual forms of activism, making Instagram a central tool for activism on
social media in the way Twitter for instance has been perceived already for longer,
additionally contributes to the timeliness of this research.

Black Activism and Racial Justice in Online Social Change Processes

Racism remains one of the major sources of inequality and violence in today’s world, deeply
rooted in our societal constructs based on a colonialist tradition. While main ways to
collectively protest oppressing systems was and still is for a big part offline action such as
demonstrating, social media has also come to play a crucial role in anti-racist activism. The
Black Lives Matter movement, started in 2013 by three Black women activists in America,

8
Alicia Garza, Patrisse Cullors, and Opal Tometi (Freelon et al., 2016; Wilkins et al., 2019)
is a key milestone in Black online activism, but its sources can be tied back to the 1960s
Black freedom movement and Black Campus Movement (Reynolds & Mayweather, 2017).
The BLM movement thus has become a widely recognized and known social movement
of the 21st century across Western countries, but it has its roots in previous work of anti-
racist activists of the 20th century. A major difference between the BLM movement and
previous activism against racial oppression though is its structure built on connective action
(Bennett & Segerberg, 2012) spread through the online sphere across space and time. The
BLM movement has resulted in offline protests, demonstrations, and other events, but it
remains mostly an online movement and connects people across the globe. For this reason,
while social movements of our time are not completely detached from the work of previous
activists, they must be analysed taking their specific characteristics such as the role of the
online environment into consideration to improve societal participation and advocacy
accordingly. While “Black Lives Matter” can refer to a number of elements related to the
movement, in this research, all references to “Black Lives Matter” or “BLM” will refer to
the online activism of the movement in general, unless stated otherwise.
The Black Lives Matter movement and the hashtags associated to it (mainly #BLM and
#blacklivesmatter) began as a way to raise awareness of racial injustice as a consequence of
the murder of Trayvon Martin (Duvall & Heckemeyer, 2018). Since then, they have been
widely used on different social media platforms by people all over the world to unite for
racial justice. The BLM movement and content related to it regularly resurface as bigger or
smaller participatory waves. For instance, in 2020, the BLM movement peaked again
globally, both online and offline as a consequence of the murder of Black American George
Floyd. This characteristic makes BLM unique as a social movement, as opposed to many
social movements centred on one specific event which determines the duration of the
movement both online and offline, BLM persists and adapts itself to the current societal
context as long as racial justice is not achieved. In the online sphere, this is seen by the
“virality” of the hashtags associated to BLM recurring, while for many other social
movements the hashtags will typically peak once, then lose visibility and use over time.
As the hashtags associated to the BLM movement have rapidly spread on the internet and
become widely used in a variety of contexts, it could be argued also that these constitute
rather memetic signifiers of support or participation rather than direct action. Gerbaudo
(2015) considers memetic signifiers to be symbolic references to a movement that constitute
elements of collective identification because of their inclusive content post-ideological

9
content and their capacity spread extremely rapidly. By analysing the process of collective
identity formation in images related to the BLM movement, this research therefore aims to
also shed a light on the true role of these images and the hashtag #BlackLivesMatter for
action against injustice.
Although BLM has become a somewhat emblematic movement of Black activism in the
21st century, it has also been widely contested, giving rise to counter-movements such as
#BlueLivesMatter, #AllLivesMatter or #WhiteLivesMatter (Wilkins et al., 2019).
Additionally, Black Lives Matter has also received its dose of critique from anti-racist
activist, for instance for reinforcing existing power inequalities it tries to counter by placing
whiteness at the centre and creating oppositions (Wilkins et al., 2019), and for being
focused mostly on Black males with its focus points on police brutality and
overrepresentation of black people in the criminal justice system (Duvall & Heckemeyer,
2018). The focus on Black males in BLM, which much anti-racist activism in general has
been criticized for (eg. Crenshaw, 1991), asks for more inclusive social activism and
research of it to overcome this gap.

Intersectionality and Black (Cyber)feminism in Social Change

The development of digital technologies and the nature of social change activism in the 21st
century has called for new and updated sociological frameworks of analysis. While women
and marginalized groups have played key roles in anti-racist activism for long, their voices
have often been silenced (Jackson, 2016). New technologies in the 21st century have brought
the hope of a more inclusive participation and the opportunity for marginalized voices to
be heard and included in the discussion. Instagram has been widely used for feminist action
(Mahoney, 2019), and women of colour particularly have also adopted social and other
new media as a tool for activism (eg. Knight Steele, 2021), which calls for a further analysis
of how Black feminism manifests itself in women’s social media activism. Several online
social movements have attempted to bring a more intersectional approach to Black
activism, one being the #SayHerName campaign, which has manifested in the form of
micro mobilizations around events of violence towards by Black women, aimed at raising
racism faced by Black women to the forefront (Brown et al., 2017). Through an
intersectional approach, #SayHerName presents a movement of Black feminism (Brown
et al., 2017) that aims to counter the downfalls of mainstream anti-racism, mostly focused
on Black men including for a big part the BLM movement, and mainstream feminism,
focused on white women (Crenshaw, 1991).

10
Intersectionality presents a framework emphasizing the role of identities in power dynamics
brought forward by Black feminist scholars. From this perspective, the study of
representations as a tool for shaping discourses in connection to identities of individuals
and groups is particularly relevant in research based on intersectionality theory. While
intersectionality doesn’t have one unique established definition, at its core is the
acknowledgement that several areas of one’s identity intersect and form a complex system
of inequalities (Collins, 2015), and for this reason, race, class, gender, sexuality, ethnicity,
and age cannot be analysed independently of one another. Intersectionality theory aims to
de-centre privilege and power structures, taking into consideration more holistically the
complex reality behind systems of oppression and inequality. The term was coined by
Crenshaw in 1991, and Crenshaw’s work remains today a key reference in intersectionality
theory. Nevertheless, the development of the concept of intersectionality per se cannot be
attributed solely to Crenshaw but was advanced rather by a whole group of scholars of
colour and still is today. As a relatively new theory that aims to take the complexity of society
into account, the framework of intersectionality changes and develops a lot, and for this
reason the lack of one unique, set definition for the term is positive as it allows the
conceptualization to be adapted to the needs in specific cases (Collins, 2015).
Overtime, intersectionality has become a key term in anti-racist discussion, and while BLM
is still criticised for being strongly focused on male-dominant topics, the decentralized
characteristic of the movement makes it possible for different groups to steer the discussion
in different directions (Ince et al., 2017). Through this, especially feminist activists have also
brought a variety of topics closer related to racial injustice more broadly and/or issues more
frequently faced by Black women. Different approaches to social media movements have
been facilitated to a different extent by various platforms. Instagram as a platform is
especially prone to women’s activism and has become an important platform for women’s
political activism (Sheldon & Bryant, 2016; Einwohner & Rochford, 2019).
Intersectionality in turn is particularly important to Black feminist activism, as Black women
are simultaneously forced to fight against several systems of oppression, including but not
limited to race and sex (Brown et al., 2017).
It is no coincidence that intersectionality theory was advanced specifically by Black feminist
scholars – while feminist theory was very much centred on white women and anti-racist
research was based strongly on inequalities faced by Black men, Black women felt that
neither approach fully encompasses the scale of inequalities they faced. Additionally, in
previous research Black women and women of colour had rather been subjects and not

11
actors of research, which further contributed to existing inequalities (Alexander-Floyd,
2012). Intersectionality has therefore played a key role in empowering Black women and
feminists of colour by providing a framework taking into consideration the superposition
of several inequalities, but also by giving them agency and voice in social research. Crenshaw
(1991) presented three aspects of intersectionality: theoretical, political, and
representational. This framework pinpoints that intersectionality is not only a theory, but a
tool of empowerment and action.
A second approach emphasizing specifically the role of digital media in social change is
Black cyberfeminism, which is based on Black feminist theory and feminist technology
studies to form an understanding of how inequalities are translated in the digital space
(McMillan Cottom, 2016). Cyberfeminism has a focus on how digital technologies enable
women to engage and participate in societal processes, be they social, economic, or
political. Black cyberfeminism brings an intersectional approach to cyberfeminism through
the analysis of mechanisms behind the ways of different groups to engage in the digital
(McMillan Cottom, 2016). Black cyberfeminism has three main themes: social structural
oppression of technology and virtual spaces; intersecting oppressions experienced in virtual
spaces; and the distinctness of virtual feminism (McMillan Cottom, 2016). These three
themes mean that structural oppression is present also in the digital sphere and technology
plays a role in how they manifest, and that virtual feminism has something distinctive.
The growing importance of intersectionality theory and Black cyberfeminism in academia
displays the need to bring more attention to marginalized groups. It has been brought up
that Internet-enabled action allows members of disadvantaged groups to regulate social
identities and social movements to bring the focus of the discussion on certain topics with
the help of rhetorical and identity strategies (Wilkins et al., 2019). In a movement such as
Black Lives Matter, who is represented and included in the movement contributes to the
formation of the framework and goals of such movement. An analysis of the representation
of women of colour in this movement thus can give information on the position of Black
women in online social activism and on how women are using a hashtag to represent
themselves on Instagram.

12
Theoretical Framework – A Collective Identity Building Approach to the
Representation of Women in the Black Lives Matter Movement

This research focuses on the analysis of the representation of women in the Black Lives
Matter movement on Instagram. The research is carried out based on the exploration of
collective identity building as a part of social movement theory, as well as a critical approach
to the technological ground and underlying commercial logics of Instagram as factors in
online activism. The importance of visuals in collective identity formation has been noted
in previous research (eg. Gerbaudo, 2015), and Instagram has been used by women for
feminist action to “navigate the neoliberal visual economy” (Mahoney, 2019), but further
research on collective identity formation on a visual-based platform like Instagram is
needed. The research aims to answer the following research question:
RQ: How is Instagram used to represent women as a part of the
Black Lives Matter movement?
Women, and especially women of colour, are key players in BLM and other activism for
racial justice. Because of the growing importance of Black feminist activism, the long
sustained marginalized position of women of colour in both social activism and society at
large, as well as its particular relevance on female-dominated platforms such as Instagram,
the analysis of the representation of women in BLM activism is needed in today’s research.
Such an approach based on empowerment of minorities and the acknowledgement of the
fact that all individuals and groups to be recognized and respected is nevertheless often in
contradiction with the neoliberal development discourse, which emphasizes the role of
market logics in who is heard, respected, and given the room to act (Tufte, 2017). This
neoliberal approach on the other hand is very tightly connected to Instagram and other
social media platforms’ commercial basis and structure around algorithms revolving around
the attention economy. Additionally, scholars have pointed out the existence of racial,
gendered and political biases in algorithms (e.g. Tanksley, 2019; Noble, 2018) – unlike
popular belief considers it to be, technology is not neutral, which calls for a critical analysis
of its role in publications published by users on platforms and the effect it has on social
change processes. This research aims to bring forward the representation of women in the
context of social media activism on Instagram and analyse the role of these representations
for social activism, taking into consideration commercial platform logics in such
publications. Instagram as a platform was selected specifically as a platform cantered on
visuals and favoured by a certain group of users with common interests in the topic of racial

13
injustice, namely women. These play a role in the nature of collective identity formation,
which is an important element of collective societal action that the research aims to explore
further. The research question will be answered with the support of two sub-questions:
SQ1: How do publications representing women contribute to online
activism against racism?
SQ2: How does the technological ground and commerciality of the
platform affect collective identity building?

Methodology
In the following section, I present the methodology on which the analysis I carry on is
based. I first present the process of data collection, and then the methodological tool used
for data analysis, including discourse analysis, cultural analytics, and content analysis.

Data Collection

The data used in this research was collected on the 11th and the 12th of April 2021 using
Python code and a package aimed for data scraping on Instagram, Instagram-scraper. The
dates of data collection were selected at random within the timeframe of this project. As
trends in social media content are unpredictable and the hashtag #BlackLivesMatter has
become established in internet users’ daily lives, a random sample is the most representative
of the environment researched and of the content published under the hashtag on a
continuous basis. The data was collected specifically on two consecutive dates to ensure a
minimal number of external variables affecting content within the timeframe of data
collection. The dataset consists of 1411 Instagram publications published on the days of
data collection including the hashtag #BlackLivesMatter. The dataset was narrowed down
to only include still images. The method allowed to collect a large dataset of images but was
limited by the Instagram API and the amount of data that can be scraped with the tools
available. Therefore, a limitation in the data collection is that it could not be determined
what proportion of the totality of images published in the days of data collection, or which
images were included in the dataset. Nevertheless, a sample of over 1400 images in the time
of two days can be considered substantial enough for the purpose of this research, as the
chosen methods of analysis rather focus on identification of patterns across the dataset
rather than in-depth analysis of individual images that would be representative of the dataset

14
as a whole. In future research, a more extensive dataset over a longer period of time can
complement the analysis started in this research.

Data Analysis

The data will be analysed in two parts using cultural analytics and content analysis, the
interpretation of which will be supported by tools based on Foucauldian discourse analysis.
The methods and framework of analysis are presented below.

Framework of Analysis

Social media publications have been studied previously both using quantitative and
qualitative tools of analysis. As both methods have their strengths and pitfalls, this research
will aim, through a mixed methodology combining both qualitative and quantitative
methods, to contribute to minimizing the limitations of both methods. As defined by
Creswell (2015), a mixed methodology consists in the combinations of both quantitative
and qualitative data collection and analysis to make the most out of the strengths of both
methods. This research therefore uses a large dataset of several hundreds of images, to
perform an explorative analysis of rather qualitative nature to identify patterns and
characteristics in the content of images analysed. To do so, the data will be analysed in two
stages using cultural analytics and content analysis, both of which are presented in the
section below. These two methods used will be complemented by tools of discourse
analysis. As conceptualized by Foucault (1979), discourse and power are strongly
interrelated, and discourse analysis can give important information about power relations
and dynamics by taking into consideration the role of social institutions in human
subjectivity and the formation of knowledge and a perception of “truth”. According to
Foucault, knowledge is not objective and value-free, which inevitably links knowledge to
power (Cheek, 2008). Discourse is therefore a way of representing reality from a subjective
point of view, and because of power inequalities, certain representations have higher power,
allowing certain knowledge to be shared above other (Cheek, 2008). Such power dynamics
are central to the research question, as they contribute to the representations of women on
social media and shape the way people represent themselves and others taking into
consideration the societal framework within which the content is produced, and which
representations and knowledge is brought forward, and which isn’t. This includes centrally
in this research the platform and the power it has in defining the nature of content, as well
as broader societal structures contributing to the representation of women and more

15
specifically women of colour. Discourse analysis therefore facilitates analysing the discourse
brought forward by a marginalized group, namely women of colour, in the social context of
Instagram, to consider how these discourses contribute to a collective identity formation
and comply or not with the norms set for visual discourse by the platform’s structure.
Additionally discourse analysis allows to take into consideration both discursive and
enacted processes of collective identity construction (Khazraee & Novak, 2018) in the
analysis of images. Rose’s (2016) “discourse analysis I”, which particularly focuses on the
dynamic between power and knowledge as tools for forging truth will be used as a reference
framework for the discourse analysis.
Social movements in general are considered to frame social issues by constructing meanings
to the audience that make them understand the situation in a certain way (Ince et al., 2017).
“Movement framing” as introduced by Snow et al. (1986) allows analysing social
movements from the perspective of how movements are presented by activists and what
they wish to communicate about the movement to the audience. This is strongly tied to
Foucault’s perception of representation and power – activists use their knowledge to
construct a certain representation of reality that advances the movement. Social media is
today used a lot for movement framing to reach a large audience, but the decentralized
nature of social media at the same time allows anyone to bring their own meanings to the
messaging on the movement (Ince et al., 2017), which adds a layer of complexity to the
analysis of social movements online. In the context of social media activism, this allows for
users to not only engage directly in the processes of social change, but also construct a
collective identity that facilitates social movements (Einwohner & Rochford, 2019) and can
empower minority or disadvantaged groups to bring the focus of the discussion on certain
topics (Wilkins et al., 2019).

Cultural Analytics

Cultural analytics is a quali-quantitative method introduced by Lev Manovich in 2007,

consisting in the use of digital methods to collect large datasets and visualizing them to
perform a qualitative analysis of the quantitative dataset and provide new ways to look at
large, cultural datasets (Software Studies Initiative, 2014; Manovich, 2017). The use of
cultural analytics for this research gives a possibility to get an overview of the dataset and
identify patterns in its content.
The analysis of the dataset was performed with ImageSorter, a data visualization software
which allows to classify images according to the dominant colour in them. A classification

16
according to dominant colour was selected as this allows to visualise a large dataset of images
easily to recognize similarities through a compositional perspective. Dominant colour can
for instance give indications on whether images are photographs or illustrations, or whether
they feature people, which give a good basis for a preliminary classification and overview of
the dataset. While this classification does not give a perfectly accurate result, it facilitates a
first exploration of the images published under the hashtag #BlackLivesMatter on
Instagram in the period of data collection, which serves as basis for the content analysis.
ImageSorter as a software was suitable for the analysis as it has been developed for this
specific purpose, it allows including a large dataset and performs the classification
automatically, which doesn’t therefore require extensive programming knowledge unlike
many alternative methods. Additionally, the software is freely available for download.
After a first explorative analysis of the dataset as a whole to get an impression of the content
published under the hashtag #BlackLivesMatter on Instagram, a second visualization was
made of only images representing women. The dataset was narrowed down to include only
any image including the representation of women, which resulted in a dataset of 354 images
(25,1% of the totality of the images collected) and ran again in ImageSorter for an overview
of this new dataset. Images of women were defined as any image including at least one
person identifiable as a woman.

Content Analysis

The second part of the analysis aims to delve more in depth in the representation of women
through a qualitative analysis of all images of women published in the two days of data
collection. A deductive approach to qualitative content analysis was selected for
classification of the content (Elo et al., 2014), and the categories of the content analysis were
selected based on the patterns identified in the visualization of the dataset on ImageSorter
as well as the analysis of the dataset by the researcher, to determine further the relation of
the images representing women to the social movement and the purpose of the content to
answer the second sub-question. Therefore, four categories were determined for
classification based on previous research on social media movements: content directly
related to the Black Lives Matter movement, content explicitly bringing forward black
women within the Black Lives Matter movement, and content related to identity and (self-
)representation. This classification was decided upon to be able to include a dimension of
analysis that takes into consideration not only social media activism as direct action for a
certain cause, but also collective identity building as a part of a social movement (Einwohner

17
& Rochford, 2019) and focus of a social movement on topics relevant to minorities (Wilkins
et al., 2019). The codebook used for classification of images can be found in Appendix 1.

Results

Cultural Analytics

The first part of the research consisted in a quali-quantitative exploration of the dataset
using cultural analytics. A first analysis was done on the whole dataset of 1411 images
collected for the research to explore the content published under #BlackLivesMatter
during the period of data collection as a whole and to use as point of comparison for the
analysis of images representing women. To take the analysis a step further and answer the
research questions, a second dataset containing only images representing women was
filtered from the totality of images used in the research. This resulted in a dataset of 354
images. These images were selected as containing at least one female figure, but it is to be
noted that this selection process was limited by the subjectivity of the researcher in
determining female figures according to physical features identifiable on the images, as well
as a binary male/female division of genders.

Complete Dataset

As a tool based on identification of dominant colours, cultural analytics with ImageSorter

can give indications on patterns in images based on repetitions of similar colours. The
analysis of the dataset (Figure 1) shows images grouped according to dominant colours. A
first overview of the dataset shows that a majority of images are based on neutral colours,
with small clusters of brightly coloured images standing out from this. Looking at the
content of images with different dominant colours, publications can be broadly classified
according to the patterns in content that come off from the clusters of colours. Firstly,
images with dominant black of white are mostly based on textual content. Many such
publications are screenshots from other social media platforms such as Twitter or
YouTube, magazine articles, or Instagram publications made in a textual form such as
infographics, memes or manifestos. Nevertheless, a part of the dominantly black or white
images also includes visual content, such as black and white or particularly bright or dark
portraits for instance. Additionally, a small number of images with bright dominant colours,
such as red, yellow, purple, or blue can also be counted in images based on textual content.
Secondly, a major part of images represents people. Such images are composed for the

18
biggest part of neutral colours such as grey, beige, brown, black, or white. Finally, it is
important to mention that some images can’t be specifically clustered according to
dominant colours but are rather spread out across the dataset, which contributes to making
the above presented classification of image type according to colour merely indicative. Such
images are for example images of objects, such as clothing or art pieces.
It can also be noted that the dataset contains a notable number of duplicates, which can
also be seen in the analysis as identical images are clustered together. Particularly, there are
many duplicates of images of Daunte Wright, whose killing had just happened on the 11th
of April and first day of data collection, giving room to a number of publications taking a
stance on the event.
Overall, this first level of analysis displays the variety of the content of images in the dataset.
Some patterns can be identified by classifying content according to dominant colour, but
this method remains broad and can rather give a first overview of the content of images
under #BlackLivesMatter at the time of data collection.

Figure 1. Visualization of the dataset according to dominant colour (ImageSorter)

Images of Women

The analysis (Figure 2) displays results similar to the one of the initial dataset (Figure 1):
most publications are composed of images with neutral colours as dominant, with small
clusters of brightly coloured images, mainly blue and red. A certain number of duplicate
images can also be identified, although no particularly large set of duplicates surfaces

19
specifically, as well as images clearly related to a same context, such as images with a similar
background or representing the same people, but with slight variations in the content.
Overall, the dataset contains a variety of images from group pictures to selfies or other
portraits. Additionally, some publications contain also textual elements. While patterns
can be identified through the exploration of the dataset, not many conclusions can be drawn
specifically based on the classification of the images according to dominant colour. It can
nevertheless give an overview of the content of the dataset and indicates for instance
different types of content existing under the hashtag, but answering the research questions
requires a closer look at the content of images which will be done in the next part in the
content analysis. Additionally, while images containing textual elements need to be analysed
further to determine their role in the Black Lives Matter movement, the noticeable number
of selfies and portraits gives a first indication of the use of Instagram for self-representation
and (collective) identity building through performative publications (Eindwohner &
Rochford, 2019) with their focus on individuals rather than the movement. This argument
is further analysed in the content analysis.

Figure 2. Visualization of the images of women according to dominant colour (ImageSorter)

Content Analysis

In the content analysis, the dataset including images of women was classified in four
categories: content emphasizing the position of black women in BLM, content otherwise
related to BLM, content related to identity building and completely unrelated content (see

20
Appendix 1). The vast majority of images, 71,75%, were related to identity building and
performative (self-)representation on social media. This included a variety of images such
as portraits, group pictures, and selfies, mostly of women of colour. Only 9,89% of images
were in any way directly related to BLM. These images presented a direct link to the BLM
movement or its topics of advocacy, for the biggest part in the form of text included in the
image, or clear images of offline protests. While two categories were established for the
analysis depending on whether such images specifically put women forward in the images,
all images related to the BLM movement were found to emphasize the position of women,
through solely representing women, positioning women at the forefront of the image
through the composition of the image, or by mentioning topics of concern specific to
women. A little under a fifth of the content, 18,36%, was completely unrelated to women
and the Black Lives Matter movement (Table 1). These publications included for instance
advertisements, pictures of artwork, closeups on clothing and screenshots unrelated to the
topic of BLM or women’s representation on social media.

Table 1
Percentual number of images in each category of the content analysis
Category 1 2 3 4
Percentage of 9,89% 0% 71,75% 18,36%
publications
included
Absolute 35 0 254 65
number of
images

Discussion

Instagram – The Challenges of a Commercially Based Online Environment for

Social Activism and Collective Identity Formation

This research consisted in analysing the framing and representation of women in the
context of the Black Lives Matter movement to contribute to existing research on societal
participation online. Overall, the results show that content associated to the hashtag

21
#BlackLivesMatter is very varied – while a part of publications can be identified as directly
related to advocacy for racial justice and the Black Lives Matter movement, this is not the
case for all images. This discussion aims to present potential reasons for this diversity in the
content, as well as the effects of it on the Black Lives Matter movement and the
representation of women in the context of a social movement online, to answer the research
question: How is Instagram used to represent women as a part of the Black Lives Matter
movement?
A quarter of the totality of the dataset contained representations of women, and for the
biggest part women of colour. This “market share” taken by women in the dataset and
under the hashtag in general and can be considered quite significant taken into account the
broadness of the content published under the hashtag, but an analysis of the representation
of men would be required for a comparative approach to the division of attention on men
and women. Wilkins and colleagues (2019) brought up that online action can be a tool for
disadvantaged groups to bring the focus of social movements on specific topics with
rhetorical and identity strategies, which is also in line with Snow and colleagues’ (1986)
theory of movement framing. A broader representation of women of colour under the
BLM movement’s publications on Instagram could be considered a way for them to actively
bring themselves forward in the movement to be included in it, and for issues concerning
them to be acknowledged by contributing to the share of attention gained by this group of
users online. Additionally, as all content directly related to the BLM movement containing
representations of women clearly puts women forward in these images, this could be seen
as a direct way to aim to emphasize the role of women in this movement that has previously
been criticized for heavily focusing on Black men’s issues (Duvall & Heckemeyer, 2018),
and steer the discussion towards the role of another group (Ince et al., 2017), namely here
Black women and women of colour. This presents an example of the role of Black feminist
thought in social media activism (Collins, 2000), and how women of colour contribute to
putting themselves in a role of actors in social change, rather than subjects as previous
research had found them often placed as (Alexander-Floyd, 2012). While BLM is not in
itself a movement built on intersectionality, the active participation and putting forward of
Black women and women of colour in the publications reveals an intersectional approach
to the movement as Instagram users are challenged to see BLM as a movement not only
related to race but entangled with other societal issues such as ones related to gender. In
the same way, the active role of women of colour on social media in the scope of the BLM
movement can be related to cyberfeminism, as social media allow women to engage in

22
societal processes and shape the way they participate (McMillan Cottom, 2016). Previous
research has demonstrated the use of Instagram for feminist activism (Mahoney, 2019) as
well as the role of digital media for Black activism (Knight Steele, 2021). This research
confirmed through the strong representation of women in BLM publications the relevance
of Instagram for black activism.
At the same time as social media give an opportunity for a freer and broader participation
of women and especially women of colour in societal processes, the digital space is not
neutral, and it is important to note how the technological structure of the platform can
influence this participation and whether existing societal inequalities are also translated into
the digital space (McMillan Cottom, 2016). For instance, Instagram has been built by
predominantly white, mostly males, and while technology itself could be argued to be
ultimately neutral, it is biased by those building it (Cheney-Lippold, 2017; Tanksley, 2019).
Therefore, while Black women and women of colour can gain an active, participatory role
in social change online and shape the discourse of the topic, the platform is ultimately not
built for them, their use of it and their needs. In some cases, it has been brought up that
not only are algorithms not built for Black women, but also against them, as content by
Black women is more likely to be erased, monitored, or flagged as hate speech (Noble &
Tynes, 2016), and reports of hate speech or inappropriate content by Black women are less
likely to be taken into consideration (Noble, 2018). This can put them in a disadvantaged
position through underlying structures and mechanisms that the platform is built on, based
on the politics set by the company (Srnicek, 2017). The importance of power in the
discourse online, but also in the underlying technological structures, contribute to changing
or maintaining the existing societal power dynamics. Hence, while Foucault (1979) argues
that representation has a decisive role in power dynamics, this argument misses the position
of the support of discourse, such as the technological structure of a platform in the case of
social media. Nevertheless, using Foucault’s framework of discourse analysis, the central
role of women in all publications related to BLM contributes to building a narrative that
empowers women and puts them in a central position in the larger narrative of racial
injustice. As social media is a space where users compete for the attention of others, this is
particularly important as these representations have more power than different narratives if
they gain a big enough attention share. This research therefore also supports the previously
presented argument that Instagram is a platform favoured by women for political activism
(Sheldon & Bryant, 2016; Einwohner & Rochford, 2019), as women tend to put their
representations forward in the scope of the hashtag used. Foucault saw that wherever there

23
is power, there is also resistance (Rose, 2016) – feminist, and particularly Black feminist
action online can very much be seen as such a resistance against a white, male-centric
construction of society which inevitably results in power and privilege for those based on
whom discourses are made. In the case of BLM, the focus on topics particularly relevant
for white males remains dominant, until contested strongly enough by competing
discourses. In this aspect, the participatory nature of social media has given more room for
alternative discourses to gain more margin and deconstruct default power dynamics.
The hashtag #BlackLivesMatter was originally created to raise awareness and mobilize
people against racial injustice. Regardless, the hashtag remains free to use for all and while
it has a use and meaning commonly accepted and understood by social media users, it can
also be associated to publications on different topics. In the images representing women, a
part of the images is explicitly connected to the Black Lives Matter movement, including
for instance textual elements related to it, or images of protesters. At the same time, a large
proportion of the publications is strongly focused on portraits and representations centred
around individuals and groups that can’t be directly connected to the Black Lives Matter
movement based on the content of the images. Many of these images could be identified
as images contributing to a certain act of identity-building on the social network: portraits
and (self)representations of women bring them forward as individuals on the platform.
While this research is not extensive enough to determine what is the indirect connection of
these images to the Black Lives Matter movement if any, with an intertextual reading of the
dataset in connection to the meanings generally associated to the hashtag
#BlackLivesMatter, it could be considered that simply the use of the hashtag associated to
the movement is an action that aims to connect the images to the cause and to one another.
Such a reading relies of the Foucauldian approach of meaning in discourse based on social
construction: the BLM movement has managed, even facing the challenges posed by
scattering of discourse online, to become known as a movement against racial injustice by
most people. The discourse associated to the movement itself thus has succeeded in gaining
a strong enough power position for the discourse on BLM and the identifiers such as
hashtags associated to it to evoke specific meanings. From this lens, the use of such a hashtag
under images could be a part of a social performance aiming to connect the individual to
the network and the movement through social identity building. This type of identity
building in turn can shape the movement and who are seen as the figures at its forefront.
While a connective movement (Bennett & Segerberg, 2012) such as Black Lives Matter
does not have explicit leaders and hierarchy in organizing it, its participants forge the

24
perception of who is leading it. Especially as a movement strongly built on online action,
the figures represented in publications related to it strongly influence the public opinion of
the movement. Such online identity building and representation of women in connection
to content related to the movement can be perceived as a way for women to empower
themselves to be perceived as participants of the movement, while expressing support to it.
From Einwohner and Rochford’s (2019) angle, such representations associated to the
movement could be seen as indirect stances around the movement, and particularly a way
to amplify Black women’s voices that have previously been unheard, but rather as a group
of individuals displaying a collective identity than through direct action in the movement.
As the researchers note, political action on Instagram can be perceived as “gendered
political performances” (Einwohner & Rochford, 2019), which can be used to shape the
discourse around the movement in a certain direction or position it according to the group
of individuals (Wilkins et al., 2019). Being a platform based on visuals and favoured by
women, Instagram sets a favourable ground for collective identity construction for women,
which can partially explain the high number of publications related to collective identity
building. Simultaneously, this can also risk diminishing the power of the social movement
by undermining content directly related to advocacy work. Hashtags work as a repository
for indexing topic-specific content and can also be used for advocacy to spread information,
in which case social identity-building can render it harder for the movement to achieve its
goal. This happened when on the 2nd of June 2020, two American music executives, Jamila
Thomas and Brianna Agyemang, started a global movement of sharing a black square on
one’s Instagram profile as a sign of solidarity for Black Lives Matter. While the intention
was good, as a result a lot of the informative content on the movement momentarily
drowned under black squares as Instagram users would accompany their publications with
the hashtag #blacklivesmatter, resulting in informative content on racial injustice being
shadowed by the abundance of black squares posed as a sign of support for the movement
(Haylock, 2020). This is an important issue for the cause, as one goal of societal activism is
the diffusion of information, and in the case of BLM, many social media users consider
social media to be a source of information related to the topic (Cox, 2017). Such a situation
displays a challenge of user-participation and social media affordances – as there are no set
rules on how a certain feature, such as the hashtags, should or can be used, this can result
in a variety of outcomes depending on how users adopt said feature.
The way users adopt a feature in its turn is for a part affected by the commerciality of the
platform. While a place of social advocacy, social media platforms are also marketplaces

25
of attention for the platforms selling data to advertisers or other parties, but also for regular
users. The attention of other users is limited, and while social media can be used for a
variety of reasons, many users eventually make their content public to be seen by a certain
audience. This can encourage regular users to adopt strategies for getting the attention of
their audiences. One such strategy can be the use of hashtags to make content visible in a
diversity of newsfeeds. Hashtags are a way of indexing content according to topic, the
algorithms behind social media platforms also make hashtags influence the general visibility
of content. While these algorithmic logics are a black box the variables of which are
unknown to most and ever evolving, it is generally accepted that the use of popular or
trending hashtags can increase the attention on a publication. Whether this is the reality or
not, such an assumption of Instagram users makes that the use of the hashtag
#BlackLivesMatter, under which already over 27,4 million posts have been shared, could
also be seen as the use of a social media affordance based on commodification of attention
for personal reasons instead of ideological ones. This could partially explain the large
amount of content that cannot be directly linked to the BLM movement, as the hashtag is
free to use and could be used as a tool for visibility. At the same time, this might be relevant
also for publications related to identity-building, as these can seek to create a feeling of
community on the platform, or gain validation of oneself by the rest of users, expressed
through likes. This relates to the debate over the impact of social media activism for social
change, which sceptics have labelled as “clicktivism” or “slacktivism” (Christensen, 2011;
George & Leidner, 2019) with little or no real impact. Nevertheless, social media platforms
are also tools to build social and political structures in a broader sense (Avram et al., 2019)
and the role of media is increasingly entangled in individuals’ lives and societal institutions
(Tufte, 2017), and can therefore not be considered as fully separate from the rest of society.
While the importance of a unique publication for social change can be debated, as social
media platforms allow a massive amount of content interconnected through networks of
users, hashtags or other affordances, the mass of content and its potential for impact offline
cannot be dismissed, while hard to measure. Relating to a Foucauldian perspective, social
media platforms and platform companies could be seen as institutions similar to the state
and institutions affiliated to it. Foucault claimed that such institutions shape discourses
based on their regime of truth (Rose, 2016), which is much like what social media platforms
do through the structure and logics of functioning of platforms. Essentially, social activism
online is in a strain between complying to the regime of truth imposed by the platform and
resisting it through various means and the development of alternative discourses. The ability

26
of users to fully resist the capitalistic logics of social media platforms can nevertheless be
discussed as it forms such a strong basis for interaction online set on the politics of the
company (Srnicek, 2017).
A challenge in social activism that has become increasingly important in the last years with
the inclusion of the internet and social media platforms into societal processes is the speed
at which discussions evolve. Compared to the early years of the internet, the participatory
characteristic of social media platforms allows users to engage with each other in the digital
space and unite across space and time for a common cause quicker and easier. Such
participation renders the speed and the quantity of communication virtually almost
limitless: as any internet user can publish any content they wish on a social media platform
at any time, the discussion and connectivity on said platforms is being continuously
accelerated. Platform companies equally use this speed and quantity of content in their
algorithms and as a part of the platform mechanisms for their advantage by favouring
“trending” content and pushing it higher to their users, as more popular content in a specific
moment in its turn is more likely to increase user engagement that provides ultimately
monetary value to the company. Therefore, once again the platform company through its
choices in the technological structure of the platform play a role in what content is popular,
and in which direction discussion on a topic is taken. For instance, if the hashtag
#BlackLivesMatter is increasingly shown to users, the result of the discussion on racial
justice on a social media platform is very different from the one that would result from
increasingly showing the hashtag #AllLivesMatter which is associated to a discourse
opposing the Black Lives Matter movement.
The role of the platform in determining which content gains the most visibility is
nonetheless not the only issue related to the fast pace of online activism. Because of the
freedom of users to associate specific hashtags to any content of their choice, as the
popularity of a hashtag grows, so does the diversity in the content associated with it. This in
turn increases the risk of content unrelated to the cause in the case of a hashtag stemming
from social activism such as #BlackLivesMatter, and the general quicker fractioning of
content as contestation and polarisation emerge quicker and the topic becomes politicised.
The speed of discussion and engagement on social media has thus resulted in a very
sensitive timeframe for optimal impact of online activism. When the cause and the
identifiers associated to it are not widely known, the reach of the content is very limited, but
as it gains momentum, eventually the capitalistic mechanisms related to the attention
economy take over the message as users adopt the hashtag for the purpose of gaining

27
popularity. In the case of the BLM movement, the hashtags associated to it have remained
in use for a longer time than the ones of most social movements. This nevertheless doesn’t
guarantee continuity or consistency of the content related to it overtime. Instead, the
decentralized nature of social media platforms has resulted in an incoherent political
message in content published under #BlackLivesMatter on Instagram and other social
media, as well as the evolution of the content over the years. In the time of this research,
the publications included a proportion of content directly related to Black Lives Matter,
but a majority of it is connected to the movement only indirectly, or not at all.
The inability in this research to determine clearly how and whether the content is connected
to the Black Lives Matter movement also raises another important characteristic of online
content. Instead of having one specific aim, the convergence of varied motives and practices
of social media use have resulted in an online culture where publications can
simultaneously serve several purposes. While a publication can aim to take a stance for a
societal cause, it can also be used simultaneously for example as a tool for identifying
oneself to a community, or to gain popularity on the platform. Without necessarily ill
intentions from the part of social media users, this intersection of several motives can
contribute to further fragmenting content and distancing it from the cause, as the
publication is crafted to fit these different needs in the best possible way, which leads to
compromises in how well it can serve one specific purpose.
This research has pointed out the complexity of analysing social media content from a
societal perspective, both on methodological aspects of collecting and analysing data, and
on the role social media in reality plays in societal processes. At the same time, conducting
this research has also given me as a researcher and as a person a better understanding of
how difficult it is to draw conclusions on how social media affects society. This has made
me reflect for instance on topics such as how to counter misinformation or how to react to
and engage with politicised topics, or how to raise sceptics’ awareness of the effects of social
media on us as individuals and as a society. Because of this complexity, it is very hard to
provide bullet-proof evidence of the problems that can arise from for instance algorithmic
biases or the commercial basis of social media, and how those contribute to shaping our
behaviour. Netflix managed to bring a broader public attention on the dangers of social
media with their 2020 documentary The Social Dilemma (Orlowski, 2020). Paradoxically
though, this documentary was produced by one of the biggest streaming platforms in the
world which in no ways is neutral in the discussion and gained a wide popularity for a big
part because of the dramatization and amplification of reality and the framing choices made

28
in the discourse to raise strong emotions. As an individual, this project has made me even
more aware than before of both the pitfalls and opportunities brought by social media for
social change processes, bringing me to the core question of how to achieve truly positive
social change and navigate the complexity of the world we live in. It has also made me
reflect on my own use of social media, and what are the sometimes unconscious motives
for framing the content I publish in a certain way: is it to make a change? To identify myself
to a certain group? To gain popularity? I do not see any of these reasons necessarily as less
valid than the other, but it is important to be aware of them and to understand how through
them we are reinforcing or resisting existing power dynamics. I also realised that such
reflections require a certain openness to vulnerability with oneself. It is not easy to admit to
oneself for instant seeking validation, or that a publication that could on the surface be
meant well to defend an important cause can also have negative effects in other aspects.
This is particularly important for social media activists, to avoid polarization and situations
where internet users are pushed, for a big part because of platform logics, to simplify
discussions and oppose one another without further discussion, instead of truly opening
the virtual space for constructive conversation. The fact that not even someone with a
formal education related to these topics is in any way immune to them makes me strongly
believe in the importance of a better education on social media use and critical digital
literacy for everyone. As a social researcher, this project has also once again brought me to
one of the core struggles in the field – how to limit ones’ research to a narrow enough
question to be able to answer it within the set framework, while not simplifying reality and
omitting the ties to the research topic to a broader societal and research framework. I see
this to be virtually impossible to achieve perfectly, and this project has also brought me to
reflect on its limitations and what could be improved in future research as presented below.

Limitations

As all research, this study presents a number of limitations that have to be taken into
consideration in its results. Firstly, from a methodological perspective, the performed
analysis only allowed to look at the images in the dataset based on the classifications
performed, which omits possible alternative classifications that could lead to different
results. For this reason, this research is rather one exploration of the publications related
to the BLM movement and further research is needed to get a more complete overview of
the data. Additionally, the limited size of the dataset and its collection only over two days
limits the results. At the same time, the methods chosen do not allow in-depth analysis of

29
individual images, which could give important indications on representation of women in
future research. Future research would thus also benefit from a comparative approach
between representations of women and representations of men in BLM online, and both
a broader quantitative analysis of a bigger dataset, and a further qualitative, in-depth analysis.
Secondly, the research was specifically focused on the analysis of the published images.
This excludes all the metadata associated to images, such as captions, profiles from which
they are collected, time of publication, and other hashtags used. An analysis including for
instance captions could complete the current research by giving a more holistic view of the
content of the image and its connection or the lack of it to the BLM movement. Similarly,
knowing which profiles images come from can indicate who are really the people publishing
these images and therefore participating in the movement online. For the feasibility of the
research, I made the assumption that most images are generally self-representations, and
therefore that the number and quality of images of women directly reflects the role of
women in the movement, but this connection cannot be validated without the analysis of
profiles images come from.
Thirdly, the time-sensitivity of research on online social movements presents significant
challenges in both data collection and analysis. As the amount of content relevant to the
research topic is virtually infinite and spread over a long period of time, the sample selected
can only give indications on the nature of the content published at the time the research
was conducted, and the results may vary vastly for a sample selected years or just only days
apart. This also leads to the research not being usable as it is to extrapolate results on future
changes or developments of an online social movement. The unpredictability of social
media poses a major challenge for research, as the components leading to a certain use of
a social media platform are connected to a number of vastly unknown variables, including
user behaviour, but also for instance past and future changes in the platform infrastructure
or algorithms. At the same time, the time-sensitivity of internet research has also limitations
in the use of previous literature: as the process of academic publication can take several
years, it is very challenging to have up-to-date research that would be directly usable in new
research as such. Instead, while previous research plays a key role in progress in academic
research, it has to always be taken into consideration that a research project and the
conclusions made in it are related to the time and context where it was conducted, and
those might get outdated quicker in Internet research than in more traditional social
research. Change in the societal, economic, technological, and other contexts must thus be
critically taken into consideration when applying previous research to new one.

30
Finally, my personal biases as a researcher are important to be taken into consideration.
Being a white, middle-class, young woman from Europe, my perspectives and conclusions
are undeniably limited by my experiences. Especially when speaking of intersectionality
and black feminism, it is important to acknowledge and put forward black women and
women of colour who have built these frameworks and practices and continue to do so,
which I have aimed to do also through the literature selected. This doesn’t nevertheless
erase my subjectivity as a researcher, which has impacted the results of this project. As a
last note, both because of my own background and the general literature used in this
research, this research presents a very Western-centric approach. BLM can be considered
a movement mostly based in the US and in Europe, but racial injustice is a global topic in
the tackling of which the West cannot take a central position. It is therefore important to
note the biases of this project related to a Western-centric approach, especially in the
application of the results to different contexts.

Conclusion
The present research aimed to shed a light on how Instagram is used to represent women
as a part of the BLM movement by analysing the position of the publications in online anti-
racist activism and the importance of collective identity building. Publications under the
hashtag #BlackLivesMatter are very diverse in their content, with a broad variety in their
connections to the actual Black Lives Matter movement. A direct connection to the BLM
movement can be established in part of the images representing women, and these tend to
place women specifically at the forefront of the action which contributes to shaping the role
of women in the social movement. At the same time, a large number of publications are
rather more general (self-)representations, in which the only connection that can be
established to the social movement in this research is the use of the hashtag
#BlackLivesMatter. These publications contribute to collective identity formation
processes on the social media platform rather than direct advocacy for racial justice.
In the context of the Black Lives Matter movement, the diversity of content in publications
representing women displays different motives behind the use of the hashtag associated to
the topic. Overall, in line with previous findings about the use of Instagram by women
(Sheldon & Bryant, 2016; Einwohner & Rochford, 2019), the representation of women
seems to be more tied to collective identity building that direct advocacy. Consequently,
publications representing women can be perceived as a part of the action for the cause, but

31
also as a performance that aims to display identity and identify oneself to a certain group,
which can also contribute to empowering women. In this sense, on one hand, social media
can be perceived as a way for disadvantaged groups to take ownership of a movement and
become active members in it through the representations associated to the movement
online. As Jackson (2016) suggests, Black Lives Matter is a social movement with a certain
level of success in bringing forward an intersectional approach to social change, which is
more inclusive of different minority groups in the deconstruction of racial oppression. On
the other hand, it could also be argued that images not directly tied to the social movement
can also undermine efforts to bring forward the cause and instead spring from the need of
individuals to gather a share of attention on social media. The use of the hashtag
#BlackLivesMatter in the context of collective identity formation and the rapid spread of
the hashtag on the platform give it memetic characteristics, as the main focus of a majority
of the content is not on direct advocacy. This also displays a challenge of social media
advocacy as content becomes quickly scattered and politicized through broad and rapid
user participation. Therefore, while processes of collective identity building on social media
can empower user groups, they might at the same time play a role in weakening the direct
impact of what has started as a social movement and turn content related to it rather into a
meme.
it is hard to determine the real impact of platform logics in publications on social media,
but the representation of women is also partly affected by both the technical structure of
the platform and the commercial logic of its functioning, in addition to how users decide to
adopt them. Social media allows users to shape discourses and participate in freer ways, but
the disadvantaged position of Black women and women of colour on the platform has to
be taken into account: while Instagram can be empowering and give a seat at the table to
minority groups, the algorithmic and technological structure of the platform is not neutral,
and this can affect the participation of said minority groups to the discussion in ways that
are more or less subtle and hard to measure (Cheney-Lippold, 2017; Tanksley, 2019).
These effects could for example include publications of Black women being deleted from
the platform, which therefore would not be controlled in the dataset.
By taking Instagram features such as hashtags, users can determine the type of content that
other users are exposed to when searching with a specific keyword, which contributes, in
the case of social movements, to the general narrative built around the movement. In this
sense, social media activism empowers internet users through a participatory culture
(Jenkins, 2006), where participants can determine the direction of the discourse, and

32
representations put forward shape the power dynamics between participants (Foucault,
1979). At the same time, the scarcity of attention and the commerciality of a platform also
contribute to how users adopt specific affordances. In the case of #BlackLivesMatter, the
popularity of the hashtag can encourage users to publish under the hashtag for reasons
unrelated to the cause as a popular hashtag can give a higher exposure to publications, no
matter what their content is. Nevertheless, algorithmic logics of social media remain a black
box, and both action of users on social media and research on it can only speculate the
functioning of it, as all parameters related to the popularity of a publication can’t be
controlled.
Further research examining the underlying motives of users to publish certain type of
content can complete this research to answer what is the role of the platform and its
affordances in online societal action and how users perceive and build their role in online
activism. In parallel to a deeper understanding of how platform affordances affect user
behaviour on social media, further research on the functioning of algorithms and the
technological basis of the platforms can complete answers to the question of how
representations are shaped on social media. Such research is also constantly needed as
social media research is very time and context-sensitive and the role of social media in social
change processes evolve constantly with user behaviour, societal context, and technological
changes. The current research can conclude that the content representing women in the
context of Black Lives Matter on Instagram at the time of the research is significant in
number and presents specific characteristics, such as putting women forward in relation to
topics of concern for racial justice. This shows an active participation of women in shaping
the discourse on the platform, be that for motives of direct activism, or of collective identity
building on the platform. The algorithmic and technological structure of the platform
affects these representations as users adopt tactics to increase the share of attention they
gain from other users, such as the use of a popular hashtag or the publication of several
similar images, but much of the impact of technology on the representation of women
remains difficult to measure through individual variables as the functioning of platforms is
a secret well kept by platform companies.

33
 References
Alexander-Floyd, N. G. (2012). Disappearing Acts: Reclaiming Intersectionality in the
Social Sciences in a Post-Black Feminist Era. Feminist Formations 24(1), 1–25.
Avram, G., Hee-jeong Choi, J. , De Paoli, S., Light, A., Lyle, P. & Teli, M.
(2019). Repositioning CoDesign in the age of platform capitalism: from sharing to
caring, CoDesign, 15(3), 185–
191. https://doi.org/10.1080/15710882.2019.1638063
Bennett, W. L., & Segerberg, A. (2012). The logic of connective action: Digital media and
the personalization of contentious politics. Information, communication &
society, 15(5), 739–768.
boyd, D. (2010). Social Network Sites as Networked Publics: Affordances, Dynamics, and
Implications. In Papacharissi, Z. (Ed.), A Networked Self: Identity, Community,
and Culture on Social Network Sites (pp. 47–66). London: Routledge.
Brown, M., Ray, R., Summers, E., & Fraistat, N. (2017). #sayhername: a case study of
intersectional social media activism. Ethnic and Racial Studies, 40(11), 1831–
1846. https://doi.org/10.1080/01419870.2017.1334934
Cheek, J. (2008). Foucauldian Discourse Analysis. In The SAGE Encyclopedia of
Qualitative Research Methods. SAGE Publications, Inc.
https://doi.org/10.4135/9781412963909.n180
Cheney-Lippold, J. (2017). We Are Data. New York University Press.
Christensen, H. S. (2011). Political activities on the internet: Slacktivism or political
participation by another means? First Monday, 16, 1–10.
http://dx.doi.org/10.5210/fm.v16i2.3336
Collins, P. H. (2000). Black Feminist Thought: Knowledge, Consciousness, and the Politics
of Empowerment (10 Ed.). New York: Routledge.
th

Collins, P. H. (2015). Intersectionality’s Definitional Dilemmas. Annual Review of

Sociology 41(1), 1–20.
Cox, J. M. (2017). The source of a movement: Making the case for social media as an
informational source using Black Lives Matter. Ethnic and Racial Studies 40(11),
1847–1854.
Crenshaw, K. (1991). Mapping the Margins: Intersectionality, identity politics, and violence
against women of color. Stanford Law Review, 43(6), 1241–1299.

34
Creswell, J.W. (2015). A Concise Introduction to Mixed Methods Research. Thousand
Oaks: Sage.
Duvall, S.-S., & Heckemeyer, N. (2018). #blacklivesmatter: black celebrity hashtag activism
and the discursive formation of a social movement. Celebrity Studies, 9(3), 391–
408. https://doi.org/10.1080/19392397.2018.1440247
Einwohner, R. L., & Rochford, E. (2019). After the March: Using Instagram to perform
and sustain the Women's March. Sociological Forum 34, 1090–1111.
https://doi.org/10.1111/socf.12542
Elo, S., Kääriäinen, M., Kanste, O., Pölkki, T., Utriainen, K., & Kyngäs, H. (2014).
Qualitative Content Analysis: A Focus on Trustworthiness. SAGE Open.
https://doi.org/10.1177/2158244014522633
Foucault, M. (1979). The History of Sexuality, Volume I: An Introduction. London: Allen
Lane.
Freelon, D., McIlwain, C. D., & Clark, M. (2016). Beyond the hashtags: #Ferguson,
#Blacklivesmatter, and the online struggle for offline justice. Center for Media &
Social Impact, American University, Forthcoming.
George, J. J., & Leidner, D. E. (2019). From clicktivism to hacktivism: understanding digital
activism. Information and Organization, 29(3), 100249.
https://doi.org/10.1016/j.infoandorg.2019.04.001
Gerbaudo, P. (2015). Protest avatars as memetic signifiers: Political profile pictures and the
construction of collective identity on social media in the 2011 protest wave.
Information, Communication & Society, 18, 916-929. doi:10.1080/13691
18X.2015.1043316
Goffman, E. (1959). The Presentation of Self in Everyday Life. New York: Random House.
Haunss, S. (2015). Promise and practice in studies of social media and movements. In
Dencik, L. & Leictert, O. (Eds.). Critical perspectives on social media and protest:
Between control and emancipation (13–34). Lanham, MD: Rowman & Littlefield.
Haylock, Z. (2020, June 2). How Did #BlackoutTuesday go so wrong so fast? Vulture.
https://www.vulture.com/2020/06/blackout-tuesday-guide.html
Ince, J., Rojas, F., & Davis, C. A. (2017). The social media response to Black Lives Matter:
how twitter users interact with black lives matter through hashtag use. Ethnic and
Racial Studies, 40(11), 1814–1830.
https://doi.org/10.1080/01419870.2017.1334931

35
Jackson, S. J. (2016). (Re)imagining intersectional democracy from Black feminism to
hashtag activism. Women’s Studies in Communications, 39, 375–379.
Jenkins, H. (2006). Convergence culture: Where old and new media collide. Cambridge,
MA: MIT Press.
Kavada, A. (2015). Creating the collective: Social media, the Occupy movement and its
constitution as a collective actor. Information, Communication & Society, 18, 872-
886. doi:10.1 080/1369118X.2015.1043318
Khazraee, E., & Novak, A. N. (2018). Digitally mediated protest: Social media affordances
for collective identity construction. Social Media+ Society, 4(1), 1-14.
Knight Steele, C. (2021). When the Black lives that matter are not our own: digital Black
feminism and a dialectic of self and community. Feminist Media Studies, 1-4.
Mahoney, C. (2020). Is this what a feminist looks like? Curating the feminist self in the
neoliberal visual economy of Instagram. Feminist Media Studies, 1-17.
Manovich, L. (2017). Cultural Analytics, Social Computing, and Digital Humanities. In
Schäfer, MT. & van Es, K. (Eds.), The Datafied Society: Studying Culture
Through Data. Amsterdam: Amsterdam University Press.
McMillan Cottom, T. (2016). Black cyberfeminism: Intersectionality, institutions and
digital sociology. In Daniels, J., Gregory, K. & McMillan Cottom, T. (Eds.). Digital
Sociologies. Bristol: Policy Press.
Noble, S.U. (2018). Critical Surveillance Literacy in Social Media: Interrogating Black
Death and Dying Online. Black Camera, 9(2), 147–160.
Noble, S.U. & Tynes, B.M. (2016). The intersectional internet: Race, sex, class, and
cultures online. Bern: Peter Lang International Academic Publishers.
Orlowski, J. (Director). (2020). The Social Dilemma. [Film]. Netflix.
Pavica, S. & Bryant, K. (2016). Instagram: Motives for Its Use and Relationship to
Narcissism and Contextual Age. Computers in Human Behavior 58, 89–97.
https://doi.org/10.1016/j.chb.2015.12.059.
Reynolds, R. & Mayweather, D. (2017). Recounting racism, resistance, and repression:
Examining the experiences and # hashtag activism of college students with critical
race theory and counternarratives. The Journal of Negro Education, 86 (3), 283–
304.
Rose, G. (2016). Visual Methodologies (4th Edition). London: SAGE.
Sheldon, P. & Bryant, K. (2016). Instagram: Motives for its use and relationship to
narcissism and contextual age. Computers in Human Behavior, 58, 89–97.

36
Snow, D. A., Rochford Jr, E. B., Worden, S. K., & Benford, R. D. (1986). Frame alignment
processes, micromobilization, and movement participation. American
sociological review, 464–481.
Software Studies Initiative. (2014). Cultural analytics.
http://lab.softwarestudies.com/p/cultural-analytics.html.
Srnicek, N. (2017). Platform capitalism. Oxford: Polity.
Tanksley, T. (2019). Race, education and# BlackLivesMatter: How social media activism
shapes the educational experiences of Black college-age women (Doctoral
dissertation, UCLA).
Tufte, Thomas (2017) Communication and Social Change: A Citizen Perspective.
Cambridge: Polity.
Wellman, B. (2001). Physical place and cyberplace: The rise of personalized networking.
International Journal of Urban and Regional Research, 25 , 227-252.
doi:10.1111/1468-2427.00309
Wilkins, D. J., Livingstone, A. G., & Levine, M. (2019). Whose tweets? The rhetorical
functions of social media use in developing the Black Lives Matter movement.
British Journal of Social Psychology, 58(4), 786–805.
https://doi.org/10.1111/bjso.12318
Zulli, D. (2018). Capitalizing the look: Insights into the glance, attention economy, and
Instagram. Critical Studies in Media Communication 35 (2), 137–150.
Zulli, D. (2020). Evaluating hashtag activism: Examining the theoretical challenges and
opportunities of# BlackLivesMatter. Participations, 17(1), 197–215.

37
 Appendices

Appendix 1: Content Analysis Code Book

The following codebook presents the four categories in which images of the dataset
representing women were divided. The categories were selected based on an exploration
of the dataset to identify patterns in the content of the images according to main elements
in them and the position and framing of women represented in these images. Table 1
presents an explanation of each category. For consistency in the analysis, the categories are
hierarchical – therefore, in a situation of conflict between two categories, the image is
classified according to the higher category out of the possible classifications.
Three categories were determined for classification based on previous research on social
media movements: content directly related to the Black Lives Matter movement, content
explicitly bringing forward black women within the Black Lives Matter movement, and
content related to identity and (self-)representation. This classification was decided upon to
be able to include a dimension of analysis that takes into consideration not only social
media activism as direct action for a certain cause, but also collective identity building as a
part of a social movement (Einwohner & Rochford, 2019) and focus of a social movement
on topics relevant to minorities (Wilkins et al., 2019), which especially women of color can
be considered. To answer the research question, we therefore analyze the role of the
publications within the movement on a division on whether they are direct contributions to
the cause, and whether so with an emphasis on women or not, or if they are rather a part
of a broader act of identity building, a performative representation of the self (Goffman,
1959; Einwhoner & Rochford, 2019) or even entertainment, and whether such content is
related the social movement or not. Such an approach is particularly relevant taken into
consideration that Instagram as a platform is specifically prone to performative acts, rather
than direct action, especially by young women (Einwhoner & Rochford, 2019).

38
Table 1
Coding sheet for representation of women in the dataset
# Category Definition Example
1 Content Direct link to the Advertisement for
emphasizing the BLM movement, Black Maternal
position of black specifically framing Health Week
women in BLM women in a central
position
2 Content related to Direct link to the
BLM BLM movement in
general
3 Content related to No direct Selfie with no
performative social connection to the apparent connection
identity building on BLM movement, to BLM
social media but representation
of the self or other
as. A form of
construction of
social identity on
the platform
4 Unrelated No connection to Image of a haircut
the BLM movement
can be determined
from the image

39