Affordances and platformed visual misogyny: a call for feminist approaches in visual methods

ABSTRACT

With social media technologies, feminist perspectives have reached parts of society traditionally uninterested in or fundamentally opposed to them. While feminist activists and allies have employed technological affordances for support, belonging, and justice, the same tools are used by actors of the alt-right to gag feminist voices. As it circulates, anti-feminist content sustains heteropatriarchy and damages women beyond the symbolic by means of trolling, doxxing, and meme wars. We address this through a review of feminist visual methods applied to the analysis of imaginaries of digital gendered hate in four case studies: (1) Greta Thunberg memes in the DENY Facebook group; (2) “Fanquan Girls” meme wars in the Hong Kong Anti-Extradition Law Amendment Bill Movement; (3) visual artefacts shared under the Twitter hashtag #SisterIDoBelieveYou; and (4) cartoons of Grace Mugabe relating to presidential succession produced in seven African countries. By reflecting on the ethos behind these four cases, we identify specific benefits to be gained from working with feminist visual methods, and contour a novel phenomenon: platformed visual misogyny.

KEYWORDS:

• Visual methods
• visual misogyny
• gendered visuals
• affordances
• feminist methods

Introduction

With the proliferation of networked social media, the impact of feminist perspectives has widened, reaching parts of society traditionally uninterested in or fundamentally opposed to them (Debbie Ging and Eugenia Siapera 2019). This everyday engagement of a myriad of users with social media provides pathways for generating and experiencing the world in multiple ways (Keiichi Omura, Grant Jun Otsuki, Shiho Satsuka and Atsuro Morita 2019), i.e., worlding, that inevitably shapes how people relate to each other. While feminist activists and allies have utilised technological affordances for support, belonging, and justice (Bianca Fileborn 2017), the same tools have been employed by actors of the alt-right to distort or suppress feminist voices (Karen Desborough 2018). Scholarship has since addressed digital gendered violence and responses to it (e.g., Elisa García-Mingo, Silvia Díaz Fernández and Sergio Tomás-Forte 2022; Jessalynn Keller, Mendes Kaitlynn and Ringrose Jessica 2018; Jacqueline Ryan Vickery and Tracy Everbach 2018). In recent years, more attention, attention has been paid to the role of visuals in gender-based violence online (see Roberta Bracciale 2020; Benedicta Adokarley Lomotey 2020; Patricia Prieto-Blanco, Elisa García-Mingo and Silvia Díaz Fernández 2022).

What we term visual misogyny here describes hate expressed through the creation of gender-ideological visuals, and/or the (re)contextualisation of visuals to serve logics of hate, abuse, and propaganda. The latter is frequently achieved via the interplay of text, visuals, and creative editing processes that dehumanise, stereotype, or otherwise discredit female voices and bodies. For example, women may be visually dehumanised or devalued through use of animal imagery (Prieto-Blanco, García-Mingo, and Díaz Fernández 2022), the stereotyping of gendered practices such as selfies as vain, trivial, and sexual (Anne Burns 2015), or gendered aesthetics such as kawaii cuteness (pink hearts, glowing stars) that render feminist activists vulnerable and childlike (see Katrien Jacobs 2022). As such, the circulation of visuals across social media platforms may serve collective narratives of gendered hate and therefore needs to be subjected to close scrutiny.

This paper aims to examine how new visualities of gendered hate emerge at the intersection of social media, participatory cultures, platform logics, and affective media practices. Through a reflection on our own feminist approaches to the study of visuals, we demonstrate how gendered dynamics of opposition collectively build imaginaries of visual misogyny in four case studies: (1) Greta Thunberg memes in the DENY Facebook group; (2) “Fanquan Girls” meme wars in the Hong Kong Anti-Extradition Law Amendment Bill Movement; (3) visual artefacts shared on Twitter under the hashtag #SisterIDoBelieveYou; and (4) cartoons of Grace Mugabe relating to presidential succession produced by cartoonists from seven African countries. Drawing on these cases, we outline how feminist approaches in visual methods allow for an understanding of the ways in which visual platform affordances may facilitate or increase visibility for visual misogyny.

Theoretical context

Platform affordances and visuality

Early literature on the Internet has described digital communities as fora for increased equality, diversity, and democratic engagement. This is perhaps best illustrated through what John Downey and Natalie Fenton (2003) describe as “counter-publicity” – the notion that digital technologies allow for social groups that have hitherto been under- or misrepresented in mainstream media (e.g., women or ethnic minorities) to gain more visibility, thereby becoming “counter-publics” through digital channels. While counter-publics have found some visibility and resonance in social media platform spaces, critical social research has since shown the various ways in which individual social groups are promoted, shaped, or marginalised through pre-existing digital divides and inequalities, algorithmic effects, and online gatekeepers (within misogyny scholarship, for example, Carolina Are 2020; Sarah Banet-Weiser 2020; Vickery and Everbach 2018). These influences have allowed for gender-based violence to proliferate via online harassment, trolling, abuse, and wider narratives of sexism or anti-feminism (see, for example, Are 2020; Ging and Siapera 2019, Mendes et al. 2019; Ov Cristian; Cristian O Norocel 2022; Vickery and Everbach 2018). Sarah Banet-Weiser and Kate M Miltner (2016, 171) describe this as a “virulent strain of violence and hostility towards women in online environments,” a trend they refer to as “networked misogyny.”

Over the past decade, misogynistic narratives have been integrated into the growing visual communication on social media platforms (e.g., Douyin, Instagram, TikTok). For the research community, this heightened significance of visuals and the emerging visualities have meant that research designs and the methods applied in these need to consider new visual formats, practices, and spaces, and consequently methodologies need to be adapted. One pathway for these adaptations is through the lens of visual platform affordances, as this paper argues. Applied to the study of social media, the term “affordances” describes platform-specific influences such as the use and adaptation of platform features around anonymity, accessibility, and interactivity (e.g., liking; see Alexander Ronzhyn, Ana Sofia Cardenal and Albert Batlle Rubio 2022). While it remains a somewhat fuzzy term, platform affordances (at times also called social media affordances or technological affordances) are defined as “(…) the perceived actual or imagined properties of social media, emerging through the relation of technological, social, and contextual, that enable and constrain specific uses of the platforms” (Ronzhyn, Sofia Cardenal, and Batlle Rubio 2022, 14).

As affordances are “relational” (Maria Schreiber 2017), they shape how social groups interact, what users may be able to do within a specific platform, and their pathways towards visibility. Individual platform features may consequently both provide or deny online visibility to users and social groups (Jeffrey W Treem and Paul M Leonardi 2013; Danah boyd 2010)—including women. For example, platform affordances may affect the visibility, absence, or exclusion of girls online (Banet-Weiser 2020) and consequently facilitate vernacular practices of networked misogyny.

However, visual platform affordances (hereafter VPA) are, in themselves, comparatively under-researched. The term has been used to highlight the “specific possibilities of aesthetic expression and visual communication” that develop from platform features and practices that centre on visual expressions (Schreiber 2017, 157). Specific aesthetics, Schreiber explains, are “entangled with both the technical and iconographic affordances of the app and also with a specific audience.” In Schreiber’s example, girls posting in more publicly facing platforms such as Instagram make careful choices to produce beautiful visuals appealing to wider mainstream publics, as practices of affirmation such as liking become visible to others.

Much of the existing work on VPA also focuses on single platforms (e.g., YouTube, Alexandra Georgakopoulou 2015; Instagram; Zoe Hurley 2019). That is even though the notion of platform affordances is based on the premise that individual platforms are subject to different logics or vernaculars that allow for certain visual practices, formats, and phenomena to thrive as users interact with these environments. To illustrate, in their visual cross-platform analysis, Pearce and colleagues display different climate visualities based on the platform on which they were popular (Warren Pearce, Suay Melisa Özkula, Amanda K Greene, Lauren Teeling, Jennifer S Bansard, Janna Joceli Omena and Elaine Teixeira Rabello 2020). In doing so, they highlighted how platform vernaculars relating to visual contents and communication provided different articulations of climate change, for example more US-oriented and news-based image formats and contents on Twitter, compared to a proliferation of memes and entertainment-based contents on Facebook, and more aesthetically oriented textless visuals on Instagram.

We consequently suggest that VPA describe affordances that are rooted in visual features, (co-)produce visually anchored platform practices, and subsequently enable or constrain specific visualities. We define VPA as specific possibilities, practices, and the emerging visualities that develop from an interplay of platform features and user practices centring on visual expression. This includes forms of aesthetic expression, manipulation of indexicality (e.g., image altering, filters), detectability and readability of visual contents (e.g., in censorship or moderation), visual replicability and recreation (e.g., through memes), affective and attachment-based dimensions of visuals, the subliminal visual narration and establishment of sociocultural norms, and visibility-based mechanisms and practices in relation to visuals (e.g., image tagging). In aggregation, these affordances relate to different dimensions of visuals: image materiality, image visibility, and aesthetics.

Platformed visual misogyny

A range of works has already explored the nexus between visualities and gender relations (e.g., on Gulf/Arab women’s visual culture on Instagram: Hurley 2019; on collective narratives through Facebook photobiographies in an Iranian women’s rights campaign:; Emad Khazraee and Alison N Novak 2018; on resistance tactics in LGBTQ+ communities:; Vanessa Kitzie 2019), albeit not necessarily in relation to misogyny. The concept of platformed visual misogyny introduced here therefore builds on two conceptual premises. The first is that platform affordances are subject to gendered dimensions (see Banet-Weiser 2020; Banet-Weiser and Miltner 2016). The second is that platforms and their specific affordances can facilitate the proliferation and/or reduction of digital hate based on their specific affordances. The latter is based on Ariadna Matamoros-Fernández’s (2017, 931) description of the role of platforms in racist narratives and contents in what she terms “platformed racism”:

it (1) evokes platforms as tools for amplifying and manufacturing racist discourse both by means of users’ appropriations of their affordances and through their design and algorithmic shaping of sociability and (2) suggests a mode of governance that might be harmful for some communities, embodied in platforms’ vague policies, their moderation of content and their often arbitrary enforcement of rules.

Based on the notion of platformed racism, we describe the amplification of misogynistic discourses and their shaping through affordances and platforms’ governance processes as platformed visual misogyny. In the cases presented below, we go a step further and argue that VPA may enable or constrain different forms of visual misogyny. VPA may then both reinforce movements striving for gender equality or those spreading misogyny and hate, potentially creating regimes of visuality that sustain, if not reinforce, heteropatriarchy. These complex relationships of platformed visual misogyny can consequently be better understood through methodological approaches that are sensitive to VPA, above all where these are considered through a feminist lens.

Visual methods and feminist methodologies

While extensive visual research focuses on gender-based violence, such studies do not necessarily apply a feminist lens in their methodologies. Feminist methods are increasingly adopted in digital social research, but hitherto rarely specifically to visual methods. Feminist approaches provide alternatives to traditional models of academic knowledge production that foreground research “as rational, disembodied and objective” (Constance Dupuis, Wendy , eds Harcourt, Gaybor Jacqueline and van den Berg Karijn 2022, 4), such as semiotics (social) semiotics in the field of visual research methods. Knowledge produced by such approaches overlooks the significant influence on texts and images of the intersectional identities of those creating, (re)producing, and interpreting them. Increased awareness of the politics of knowledge production has led feminist researchers to propose alternative ways of making sense of the world that also take into account ethics, care, and responsibility in research (Ariella Azoulay 2019; Tina M Campt 2017; Marianne Lijeström 2010; Carolyn Pedwell 2010).

Although feminist approaches differ (Wendy Harcourt, Karijn van den Berg, Constance Dupuis and Jacqueline Gaybor 2022), feminist methodologies are known for granting attention to knowledge as “inscribed in the body, sense of self and community,” showing that “knowledge is not disconnected from, but rather entwined with, emotions and experiences” (Dupuis et al. 2022, 4), and addressing intersectional complexity in their analyses (Kimberlé Crenshaw 1989; Sandra Jeppesen, Toni Hounslow, Sharmeen Khan and Kamilla Petrick 2017). Feminist methodologies therefore allow for an exploration of the researcher’s role and what research practices mean for the kinds of knowledge being produced. This involves granting attention to why and for whom knowledge is sought, and how it is shared, incorporating how researchers shape research questions, their reading and citing practices, and, therefore, within which bodies of knowledge they embed themselves and their work (van den Berg Karijn and Leila Rezvani 2022).

Feminist methodologies are applicable to all social research, including digital and visual social research, because gendered power dynamics are evident across contexts. They offer pathways to understanding “the interplay between technology, users and their varied contexts” (Wendy Willems 2021, 1678) and can be employed to draw parallels between online and offline misogynistic tendencies. For instance, Keller, Mendes, and Jessica (2018) examined media texts with girls’ and women’s understandings of new media and the contexts within which they use them, which produced a comprehensive mapping of “the complex relationships between users’ multiple media engagements and their social and cultural context”, thus enabling an understanding of “digital media from the perspective of participants” (Keller, Mendes, and Jessica 2018, 24). Feminist methodologies are therefore essential in investigating the role of VPA in gender politics, acknowledging unequal power relations between genders but also allowing for consideration of visual practices as characterised by power dynamics and resistances, so that power is seen as contested rather than fixed.

Gender-based power relations have been addressed through feminist research ethics—for example, through elaborate research designs that account for iterative evaluations of consent (Patricia Prieto-Blanco 2016) and collaborations with participants that “truly promote the creation of non-hierarchical relations” (Stefano Piamontese 2021, 192), with both methods acknowledging and trusting the researcher’s competence to work with guidelines instead of rules (Lars Frers 2021). When discursive—material links between practices and subjects are acknowledged as power relationships, scholars may also decide to apply principles of reflexivity and relationality in research, to which they attach themselves (i.e., positionality). Pragmatically, this materialises into a) allowing for uncertainty to be explicit in research (Patricia Lather 2007; Corinne Squire 1995); b) combining historical contextualisation with the specificity of embodiment (Hedy Bach 2007; Pedwell 2010); c) incorporating self-reflection; and d) acknowledging individuals’ differences in order to collaborate through these (Lijeström 2010).

In what follows, we present a range of worked examples of feminist approaches through the lens of VPA. While feminist approaches are common in theoretical frameworks and empirical investigations in digital social research, we argue specifically in favour of integrating them into visual research methodologies based on platform-specific visualities. Such research allows for critical examination of visual expressions of gender dynamics in light of VPA, i.e., platformed visual misogyny.

Data and methods

A feminist approach to platformed visual misogyny

This exploration of platformed visual misogyny is an attempt to incorporate and systematise similar observations we had when working with diverse visual materials, national and media contexts. The individual projects and collective works presented here apply a feminist lens in that they a) acknowledge the ways in which researchers’ intersectional identities impact how they research; b) prioritise dialogue and collaboration in all stages of the research process; c) try to make justice to the complexity of the objects of study (within the limited space provided); and d) are targeted at open access outlets as often as those behind a pay-wall.

All four cases employed a feminist lens on a methodological level, and explored visual misogyny conceptually. We acknowledge that our individual analyses of visuals subvert customary interpretations focused merely on aesthetics and/or institutionalised meanings, and instead propose an embodied visual analysis; this is to look and interpret visuals from our guts. This strategy responds to our collective long-term experiences of engaging with images affectively instead of purely factually. This approach requires being comfortable with uncertainty as there is no predetermined scheme or steps to follow, it is more labour- and time-intensive, and it demands a constant exercise of (self-)reflection. In doing so, we demonstrate how gendered dynamics of opposition build imaginaries of gender-based violence, which are materialised in misogynistic verbo-visual outputs and respond to the logics of platformed visual misogyny.

Case study backgrounds

This paper draws on four case studies. DENY (pseudonym) is a public Facebook group (>7000 members) that positioned itself as a place for posting and discussing evidence that anthropogenic climate change is not taking place (i.e., it is settled science or a conspiracy theory). The group was observed and analysed through digital—ethnographic observation, from which a subset of visual data was separated out for identifying visual vernaculars on gender perceptions. Fanquan Girls are girls and young women aged 10–25 who worship idols, such as singers, in the entertainment industry in mainland China. During the 2019 Hong Kong Anti-Extradition Law Amendment Bill Movement, many Fanquan Girls rallied and initiated meme wars (Ally McCrow-Young and Mette Mortensen 2021), propagating pro-China ideologies. The researchers examined how images and memes were prepared, reproduced, disseminated, and used during the meme wars on different social media platforms. Grace Mugabe, the wife of Zimbabwe’s then-president Robert Mugabe, consolidated her power and became a contender in a violent presidential succession battle between 2014–2017. This study focused on cartoons about Grace Mugabe relating to this battle, produced by political cartoonists based in various African countries and published online in news publications, Facebook, and Twitter. The analytical framework combined African feminism and discourse analysis to demonstrate that sexism and misogyny are common tools for undermining powerful African women politicians across many parts of Africa. #SisterIDoBelieveYou (for other dimensions of this hashtag movement: Elisa García-Mingo and Patricia Prieto-Blanco 2023; Prieto-Blanco, García-Mingo, and Díaz Fernández 2022) encapsulates the case of believing a victim of rape in spite of a judicial sentence from 2018 that found the five perpetrators guilty of sexual abuse only, thereby implying that when the five gang-raped an intoxicated 18 year old, filmed, and photographed it, there was no violence involved. #SisterIDoBelieveYou was subsequently used for years by feminist activists and allies, whose efforts secured a change in legislation,¹ until it was hijacked by anti-feminists. Here, attention was paid to how aesthetics and design were used first by the feminist movement itself to appeal to wider audiences, and then to smooth the transition from feminism to anti-feminism. For an overview of the case studies, see Table 1.

Table 1. Case studies.

Case	Fanquan Girls	Greta and DENY	#SisterIDoBelieveYou	Grace Mugabe cartoons
Platform and platform sub-space	Facebook account, chat groups on social networking sites	Public Facebook group DENY	Twitter	Facebook, Twitter, Online news sites
Regional or cultural context	China	Global North/Anglo-Saxon	Spain LatAm	Sub-Saharan Africa
Single-, multi-, or cross-platform	Cross-platform	Single-platform	Single-platform	Multi-platform
Method of data collection	Observational, manual	Observational	Manual	Manual
Method of data analysis	Visual discourse analysis	Visual discourse analysis	Collaborative and embodied visual analysis	Visual discourse analysis
Type of visual data	Memes	Memes and cartoons	Digital images (e.g., memes, posters, photographs, illustrations, banners)	Cartoons
Comparative elements	A multi-modal comparison: visual/audio/textual	A comparison of gendered visual narratives	Hashtags; Colour; Repetition; Sharing date	Multi-modal comparison: visual/textual
Samples (timeline, sample size)	247 visuals 30.09.2019–.1.10.2019	Visuals 1.05.2019–1.06.2021	696 visuals May 2018– August 2020	49 visuals 2014–2017

The four cases aggregated here have been chosen on the basis of the different dimensions that either facilitated or constrained visual misogyny: 1) content moderation of visual gender violence; 2) critical visual creation and performance of gender identities, i.e., exploitation through stereotyping (of women, femaleness, and femininity), diminishing, branding, falsifying, reinforcing imaginaries, movement-hijacking all cases; 3) the role of aesthetics and design in gender violence; and 4) feminist research ethics in visual misogyny. In accumulation, they also represent different national contexts, suggesting that (beyond nation-specific regulation), the thriving of visual misogyny is shaped by VPA.

The datasets combined

This collaborative work emerged after the individual data collection processes. We met at the “Visual Politics and Protest” ECREA Visual Cultures pre-conference, and at the annual conference of the European Consortium for Political Research in Innsbruck (both 2022). These conversations highlighted similarities of our work in the circulation of visuals online, contemporary feminism, related reactionary movements, and boundary-pushing of established methodological approaches to the analysis of visual material. We decided to employ the differences in sampling size and research design from our research (overview in Table 1) as a framework towards showing (a) the relevance of diverse platform contexts in giving rise to visual misogyny, and (b) the consistent contribution a feminist approach generates for visual data analysis. After a process of peer-review of each other’s work, and having engaged in continuous critical dialogue through online meetings, we established that the basis of comparison across the case studies was their grounding in platform-based research, the collection and analysis of digital visual data, the application of a feminist perspective, and the topical focus on politically motivated and/or politicised visual misogyny.

One of the unreconciled differences of the data presented here across the individual case studies relates to its ethical treatment, which happened prior to our collaboration and is the result of differing research designs as well as of the contextual nature of research ethics (see AoIR Ethical Guidelines 3.0 by Aline Shakti Franzke, Anja Bechmann, Michael Zimmer, Charles Ess and the Association of Internet Researchers 2020). For a more homogeneous and ethically sensitive visualisation of these datasets, we decided to reproduce images through “image composites” (layered images, as in Pearce et al. 2020), which renders individual images almost unrecognisable and unsearchable. Thus, our collective work responds to the commonplace feminist strategy of incorporating complexity to research while attending to particularities (Pedwell 2010, 118) to contest customary modes of knowledge production.

Researching platformed visual misogyny

A feminist approach across the collective case studies revealed different ways in which individual platform affordances facilitate the spread/reproduction of visual misogyny. VPA contributed to the limiting, promoting, and/or distorting of the visibility of visual misogyny within or across platforms and their sub-spaces. In what follows, we document this process along four categories.

1) considering practices of exclusion and moderation in visual gender violence

Considerable differences across the case studies were moderation practices on individual platforms. Although misogyny was observed across all cases, the levels of gender violence displayed in visuals differed. The DENY Facebook group was the case with the highest access thresholds and consequently lowest accessibility and visibility, but also displayed the strongest visual violence. Visual misogyny was expressed through visual narratives of GT that infantilised, insulted, sexualised (then a minor), and discredited her on the basis of being female, her age, her status as a “foreigner,” and her assumed lower intellect/value (see Figure 1). These trends remained undetected outside the group due to the cohesive subspace moderation practices that removed opportunities for disagreement, as well as the relatively innocuous titling and self-description of the group as an alternative science forum. The ethnographic observation showed that the high level of visual violence was a result of two dynamics: exclusion and moderation (see Suay M Özkula 2023). Group members displaying contrary values, for example by questioning or disagreeing with posts, were trolled, blocked, and expelled from the group. This tendency created strong in-group cohesion, which meant that misogynistic visuals were not flagged, removed, or otherwise moderated.

Figure 1. DENY image composite 1. © Suay M. Özkula and Michael J. Bonner.

These trends were revealed through a feminist approach to visual data analysis, in which data were reviewed and scanned for potential gender themes. A reiterative data review of gender-based elements in visual vernaculars in subsets of data revealed persistent visual expressions of misogyny and hate towards GT. That is even though gender was not a topical focus of the DENY group as it described itself as an alternative science group and did not imply any specific linkages to gendered themes. On the surface-level, gender was therefore largely absent in verbal language and only present visually in misogynistic memes. GT (occasionally alongside other female politicians) was the target of dominant group narratives and visual practices, including framing her as a puppet or person of low intellectual capacity in cartoon and meme depictions, at times with text inserted directly in the images (see Figure 2). The reiterative data review consequently allowed for centring gendering practices in light of VPA (the platform subspace moderation of visuals).

Figure 2. DENY image composite 2. © Suay M. Özkula and Michael J. Bonner.

A reiterative data review approach was also employed in the Fanquan Girls case through a teasing out and comparing visual and textual data related to gender with different modalities and degrees of public visibility. Although the Fanquan memes indicated female gender identities, some participants in the Fanquan Girls private chat groups used male representations in their personal profiles. The reiterative interrogation of gender dynamics therefore allowed for the identification of “collective gender identity,” revealing insights into different modalities of gendering or falsifying gender (e.g., through varying visual or textual elements on Facebook, Sina Weibo, and QQ). Despite the visibility of their profiles, these gender identities were specifically chosen for gendered digital performances and largely left undetected.

2) understanding the critical visual creation and performance of gender identities

Beyond users’ specific genders, gender identities were carefully constructed and performed across the case studies. In the Grace Mugabe cartoons, a feminist approach highlighted how socio-political contexts shape visual stereotyping of gender performances and the normalising of visual misogyny across different online platforms. Grace Mugabe was displayed as a violent and unruly woman, characteristics that are in violation of the virtues of a “good mother” (see Figure 3), and as having violently manipulated and emasculated her husband, the President, to gain power (see Figure 4). A feminist approach allowed for creating constant awareness of how gendered power and domination influence the production and sharing of knowledge and discourse on women politicians. Research on the Grace Mugabe cartoons was designed specifically to investigate how gendered power dynamics in a violent succession battle were reflected in political cartoons through gender-ideological visuals that highlighted Grace Mugabe’s perceived flaws as measured against stereotypical patriarchal expectations of femaleness and womanhood.

Figure 3. Grace Mugabe image composite 1. © Norita Mdege and Michael J. Bonner.

Figure 4. Grace Mugabe image composite 2. © Norita and Michael J. Bonner.

Across all four cases, a feminist approach also facilitated the exploration of the many ways aesthetics played a significant role in how visual misogyny was expressed. The use of caricature in the Grace Mugabe and GT cartoons resulted in the exaggeration of the perceived gendered flows and transgressions in a way that made visual misogyny more acceptable as it was concealed behind humour. For instance, while representing Grace Mugabe as a witch may be humourous, it supports the patriarchal discourse in Zimbabwe that brands ambitious women politicians as witches who should not have access to power because, as Gibson Ncube (2020) notes, witches disrupt the traditional definitions of women’s roles and the patriarchal order in politics. Similarly, in the DENY group, Greta memes and cartoons edited to show her older and aggressive, taking away her role as a teenage activist and branding her an aggressor.

Visual stereotyping was also observed in the Fanquan Girls case, above all through visual-social norms that allowed for certain gendered visualities to become memetic or viral. Femaleness and femininity were employed, maintained, and exploited for political purposes via the use of aesthetics of cuteness. Media agencies conveying pro-China ideology employed images in cute manga style to depict Fanquan Girls as cute girls with big round eyes and long hair. They were constructed as young female patriotists wearing red hats with rabbit ears and attached to the China national flag and the Chinese character “fan(饭),” looking confident, cynical, or a little arrogant (see Figure 5). As such, femaleness was exploited through a visual aesthetic performance of gender identities that was rooted in stereotyping and a perceived “harmlessness” that discredited women.

Figure 5. Fanquan Girls image composite. © Xuanxuan Tan and Michael J. Bonner.

Aesthetics also filtered into visual-social norms, for example in colour choices in the feminist movement behind #SisterIDoBelieveYou, fulfilling a pedagogical function (García-Mingo and Prieto-Blanco 2023, 496–7) and reaching wider audiences. The dominant colour in the sample was dark purple, which became degraded to pink. Along with this aesthetic degradation, the discursive specificity became diluted too. Figure 6 maps the colour strategy within the feminist movement over time and illustrates how colour was intrinsically linked to the pedagogical function of the movement. Table 2 illustrates the pedagogical function and discursive dilution associated with the colour pink, which we have called “the instagramatization of #HermanaYoSíTeCreo” (Prieto-Blanco, García-Mingo, and Díaz Fernández 2022).

Figure 6. Timeline of digital visuals shared on Twitter under #SisterIDoBelieveYou. Attention to colour. © Patricia Prieto-Blanco.

Table 2. Degradation from purple to pink and from niche to mainstream feminism. © Patricia Prieto-Blanco.

The anti-feminist re-appropriation of #SisterIDoBelieveYou also applied colour-based visual-social norms, but in this case to anchor their hate to more niche feminism, as they used darker and more vibrant shades of purple and fuchsia, instead of lighter pink tones, to write “Puta Zorra Feminazi” (see Table 3). This chromatic strategy is complemented by the animalisation of feminists, the discrediting of the act of believing by a pledge to return to “scientific evidence,” and the use of formal script and calligraphic type fonts.

Table 3. Visual re-appropriation of #SisterIDoBelieveYou by anti-feminists. © Patricia Prieto-Blanco.

The collective visual repertoires across these cases reflect the critical visual creation and performance of gender identities including: how specific aesthetic choices formed expressions of visual misogyny through stereotyping, diminishing, branding, falsifying, reinforcing imaginaries, and the hijacking of feminist movements.

3) feminist research ethics in visual misogyny

Gendered experiences also became relevant in self-reflective processes, as individual scholars’ positioning, emotions, and experiences in terms of their intersectional identities were part of the research process. For instance, an analysis of visuals linked to #SisterIDoBelieveYou involved developing a collaborative protocol to be carried out by at least two researchers: one with first-hand experience of the affective dimension of the study object and another familiar with it from a distance. This protocol allowed the researchers to gain ethnographic sensibility towards the images and to sustain a continual process of self-reflection on the researchers’ positionalities and embodiments.

In comparison, a single researcher was employed in the DENY case, and positionality became relevant in terms of visual vernaculars that displayed implicit visual (and at times intersectional) misogyny. Although forum rules and moderator statements suggested that the Facebook group was an inclusive space, exclusionary patriarchal practices became visible in memetic misogyny. Questions of positionality and needs for protective mechanisms thus specifically arose from intersectional feminist readings of visual data that was not collected on the basis of gendered groups, practices, or phenomena (cf. both Fanquan Girls and DENY in Data and Methods), but where intersectional elements such as age and (dis)ability status formed part of visual misogyny narratives (cf. Figure 1). These protective mechanisms were consequently applied to both the researcher (through non-disclosure strategies) and female participants in that space (e.g., the use of image composites).

As such, different applications of positionality allowed for self-reflective processes in reading and interpreting visual data in light of our individual research identities.

Feminist methods and visual platform affordances

The feminist approaches applied to visual research across these four cases were tied to platform-specific dynamics, i.e., platformed visual misogyny. They illustrate how the adoption of a feminist lens in digital visual research allowed for an unpacking of gendered visual tactics and curation efforts, an understanding of the role of VPA in these practices, and insights into gender dynamics and data. For example, the publication of the Grace Mugabe cartoons across social media sites made them more visible both nationally and regionally, leading to intertextual interpretations in which misogynist meanings accumulate across cartoons from different geographic regions so that a cartoon from one country is understood within the context of cartoons from other countries. In comparison, the VPA of the public but moderated Facebook group DENY meant that visual misogyny was not (easily) detected by Facebook content moderators. Violence-suggesting and sexualising memes were neither flagged nor picked up because the reactionary content was communicated visually only. This included memes and cartoons sexualising GT, mocking her learning disability, and posts suggestive of physical violence (cf. Figure 1). The lack of accompanying text allowed for such images to roam undetected by automated means as these are typically designed to read textual rather than visual content. Thus, the platformed practices of this more authoritatively moderated Facebook group limited the reach of visual misogyny based on moderator preferences.

Elements of visibility tied to visual platform affordances also played a key role in the Fanquan Girls case, where different modalities of feminist elements were compared in their levels of public visibility. More feminine memes (those showing pretty and confident girls using heart and rainbow imagery) posted on Facebook accounts with a high number of fans achieved a higher level of visibility than textual conversations in a Sina Weibo private chat group blaming men disguised as Fanquan Girls. Specific platformed affordances enabled certain feminine elements to become dominant and to play a significant role in maintaining a particular form of feminine visual stereotype: Fanquan Girls as sweet, obedient, and docile patriotic fan girls who, when offended, reacted aggressively (tantrum-like).

Although the individual cases took place within separate national contexts (cf. Table 1), VPA of visibility informed both their online and offline reach. For example, in the case of #SisterIDoBelieveYou, images were shared online and offline, in a persistent back-and-forth movement that enhanced the visibility of the movement at different levels. Figure 7 shows a composite that reflects the defiance of rape culture by holding the rapists visually accountable, amalgamating visuals shared on Twitter that were also featured on other media platforms (national newspapers and television broadcasts), and displayed on store windows, bus shelters, and as old-fashioned paper posters on walls.

Figure 7. #sisteridobelieveyou image composite. © Patricia Prieto-Blanco and Michael J. Bonner.

The collective case studies illustrate the different ways in which individual platform affordances made visible elements of visual misogyny—limiting, promoting, or distorting their visibility within or across platforms and their sub-spaces. The visibility and discoverability of visual contents were shaped by technological affordances (e.g., moderation and private-public accessibility), tactics of visual curation, and tagging mechanisms such as hashtags.

Discussion and conclusion: a case for feminist approaches in visual methods

The case studies highlighted the complexities of how women are visually portrayed and exploited online, how gender stereotypes are visually normalised, and the ways in which these images unify over gendered hate. A feminist lens in method highlighted a) gendered visualities—regardless of national or media contexts, and b) underlying misogynies—whether the case studies centred on women or not. The combined case studies therefore gave rise to a range of methodological considerations for networked misogyny documented here as informed by VPA. They provided a view into the complex enmeshing of platform affordances, visualities, and gendered dynamics in what we consequently termed platformed visual misogyny. Based on the cases documented here, we define platformed visual misogyny as gendered practices and visualities that emerge from VPA, such as (in the cases here): 1) content moderation of visual gender violence; 2) the critical visual creation and performance of gender identities, i.e., exploitation through stereotyping, diminishing, branding, falsifying, reinforcing imaginaries, movement-hijacking; and 3) the role of aesthetics and design in (re)generating gender violence.

Affordances and platformed visual misogyny

Across the cases presented here, VPA became methodologically relevant in a number of ways. Through their capacity to spur, limit, or change the visibility of visuals, they both enabled and constricted specific visual discourses—either allowing for visual misogyny to become viral and thrive or for it to remain relatively under-detected. Although affordance theory is by now a rich and developed research field, these cases make several contributions to this field. As a concept, platformed visual misogyny acknowledges the complex ways in which visual sexisms and affordances intersect to create specific gendered narratives. In particular, the concept’s application here illustrates the complex intertwinement of affordances of visuality and visibility in relation to, in addition to, disentangled from, and deliberately disassociated from textual contexts.

These linkages are consequently significant for how digital-visual phenomena and practices are inscribed into research designs. In linking affordance theory to epistemological considerations, the four cases illustrate the relevance of considering VPA on a methodological level, where the ways in which platforms negotiate visualities and their visibility affect how certain contents become prominent and visible to both media audiences and the researchers studying these. While the relevance of power dynamics in research design is well known in feminist approaches to research methodologies, power influences generated by platform structures and the emerging affordances have hitherto seldom been acknowledged in similar ways (i.e., as methodological or epistemological concerns rather than theoretical approaches or empirical results). This paper therefore suggests that methodological considerations of platform affordances serve a similar purpose to feminist approaches in that they both acknowledge and interrogate the influences of these power dynamics in research design, as exemplified by platformed visual misogyny.

Methodological contributions

In terms of wider methodological implications, this paper has argued in favour of applying feminist approaches to visual analysis. Although a breadth of feminist methodological work already exists, this paper discussed a range of methodological practices that address visual research more specifically and therefore platformed visual misogyny. Three axes of feminist visual methods emerged from this enquiry, and produced relevant questions for feminist methodology:

1. Data: in what ways are visual data gender-ideological?

2. Positionality: how and for whom do we apply ethics of care?

3. Affordances: how and to what extent to these data become visible & discoverable?

Feminist visual methods based on these axes may be employed by incorporating a) considerations of ethics, care, and responsibility in visual research and data, involving accounting for researcher positionality and examining the role of the researcher for a constant awareness of power dynamics in research processes including making sense of visuals; b) sensitivity to intersectional complexities, granting particular attention to those on the margins of power to build more inclusive knowledge of platformed visual misogyny; c) attention to platform-specific power dynamics to tease out how these influence visual curation, aesthetics, and tactics of both dominant and marginalised groups; d) reiterative data reviews that allow for observation of gendered elements in research spaces where gender is decentred, falsified, or re-identified; and e) considerations of visual platform affordances towards identifying and/or contextualising platformed visual misogyny.

Future research directions

While these worked examples did not allow for a full multi-method comparative research design, they hopefully provide the groundwork for such work in the future. In particular, we hope that the methodological approaches presented here offer avenues for future visual research methods to enable more in-depth engagement with gendered visuals and visual gendering. It is hoped that subsequent research endeavours will provide further insights into these areas and engage with visual misogyny on a wider level. In particular, it would be fruitful to explore in more detail the ways in which visual misogyny is made mainstream or viral through specific VPA. Finally, further research is also needed towards examining in more detail how visual misogyny is entangled with algorithms, platform architecture, and audience practices and perceptions of media affordances in different (national, media, and other) contexts.

Acknowledgement

We collectively wish to the ECREA Visual Cultures Section for providing a platform for collaboration, Michael Bonner for the creation of the image composites, and Maria- Magdalena Bărăscu for her review comments on the paper. X.T. wishes to thank Zhishuang Wang, previously a student at Jinan University, for her assistance in the data collection for Fanquan Girls. S.M.Ö. is supported by funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement number 101027274. S.M.Ö. additionally thanks the Economic and Social Research Council for funding Making Climate Social (ES/N002016/1) through the Future Research Leaders programme, under which the initial sampling stage was funded. Norita Mdege is supported by funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement number 101032680.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The work was supported by the Economic and Social Research Council with grant number ES/N002016/1; HORIZON EUROPE Marie Sklodowska-Curie Actions [101027274 and 101032680].

Notes on contributors

Suay Melisa Özkula

Suay Melisa Özkula is a digital sociologist and post-doc at the University of Salzburg; her research interests lie at the intersection of digital activism, sustainability discourses on social media platforms, and digital and visual methods.

Patricia Prieto-Blanco

Patricia Prieto-Blanco lectures at the Sociology department, Lancaster University in the UK; her interdisciplinary and practice-oriented expertise includes visual research methods, image-based activism and media practices in the context of migration and kinship.

Xuanxuan Tan

Xuanxuan Tan is a a postdoctoral researcher at the Southern University of Science and Technology; her research interests include nationalism and activism in contemporary China, especially as it relates to fandom.

Norita Mdege

Norita Mdege is a Marie Sklodowska-Curie Research Fellow on the project “WOMPOL-AFRICA: Women Politicians in Africa: War of Symbols and the Struggle for Political Legitimacy” at the Graduate Institute of International and Development Studies.

Notes

1. In October 2021, the Organic Law of the Comprehensive Guarantee of Sexual Freedom was passed in Spain. This law situated consent in the centre, and was colloquially known as the “only-yes-is-yes” law. A subsequent reform displaced consent and saw Spanish legislation going back to evidence on the body of the victim as main consideration to interpret abuse as rape or not.