The Left Out Project Report: The case for an online gender-based violence framework inclusive of transgender, non-binary and gender-diverse experiences

The Left Out Project, supported by the Feminist Internet Research Network (FIRN), explored online gender-based violence (OGBV) as experienced by transgender, non-binary and gender-diverse (TNBGD) people in Botswana, Rwanda, South Africa and Uganda. Through this project, we aimed to centre TNBGD people’s experiences of OGBV so that responses to OGBV (e.g., policies) may be better informed and ensure stronger protections for TNBGD people.

TNBGD people are often subjected to violence based on their gender identity. Transphobic violence manifests in a number of ways, which can range from verbal to physical abuse, sexual violence, and even murder. Transgender women, in particular, experience the highest rates of violence. In 2021, Muller et al., in their study of nine African countries, reported that “three in four transgender women (73%) had experienced a form of violence in their lifetime, and almost half (45%) in the past year.” Transphobia and the threat of transphobic violence have a significant impact on transgender people’s mental health.

Violence against LGBTQIA+ people, and in particular transgender people, has increased worldwide. The spike in hate crime-related murders is a result of a “culture of violence” that emerges from transphobia and intersecting discriminations on the basis of racism, sexism and homophobia. It bears noting that police reports and crime statistics do not accurately reflect the degree of transphobic crime, due to the misgendering of transgender people by police, the criminal justice system and the media. On the African continent, LGBTQIA+ rights are in a dismal state, and as Iranti reports, “violence against LGBTQI+ people is a critical area of concern for the African continent.” Where countries do afford some state protections, these protections do not always translate from paper to people, in that high rates of homophobic and transphobic violence are still found.

Botswana is a country in Southern Africa, neighboured by South Africa, Zimbabwe, Namibia and Zambia, and forms part of the Southern African Development Community. In 2019, the High Court of Botswana repealed laws that criminalise and discriminate against LGBT people. In 2017, the High Court ruled that a transgender man was allowed to have official documents that reflected his gender identity. This judgement meant that transgender people in Botswana can now have their identities recognised and affirmed legally.

Rwanda does not outlaw being lesbian, gay, bisexual or transgender, but sexual identity is not frequently discussed or acknowledged. Rwanda is “a signatory to the 2011 United Nations statement condemning violence against LGBT people,” but, like South Africa where LGBT rights are protected, socially homophobia and transphobia persist. While the constitution protects citizens from discrimination, protections for sexual orientation and gender identity are not explicitly stated in the same way that protections based on race or sex are, for instance.

South Africa is often celebrated, both locally and globally, as the champion of LGBTQIA+ rights, with the country’s post-apartheid constitution being the first in the world to protect people regardless of sexual orientation. However, many LGBTQIA+ South Africans continue to face homophobic and transphobic harassment and violence within their communities. In July 2021, the South African Government News Agency, SAnews, reported that “South Africa has recently seen an increase in the number of attacks again lesbians, gays and transgender people,” and that 29 cases of hate crimes against the LGBTQIA+ community had been documented since the start of 2020.” Meanwhile, Gender DynamiX “documented 60 cases of human rights violations perpetrated against trans and gender diverse persons in the space of three months.”

Uganda’s LGBTQIA+ community is currently the target of the country’s Anti-Homosexuality Bill, which criminalises same-sex relationships and sexual activities between members of the same sex. This makes them the target of state violence as well as public discrimination and violence. The Ugandan government squashes political organising; for instance, in August 2022, the government banned Sexual Minorities Uganda (SMUG), a LGBTQIA+ rights organisation, from operating. There are no laws that outright criminalise transgender, non-binary and gender-diverse people, but they “have been indirectly criminalised under the offences of ‘personation’ (false representation), public indecency and the criminalisation of consensual same-sex sexual acts.”

Our research uncovered that violence affecting TNBGD people online is widespread across the four countries through systemic, cultural, religious and political influence. Our participants reported discrimination from health care professionals, being arrested and detained, sexual harassment and violence, online harassment, and being targeted by transphobic groups, to name only a few. The Left Out Project takes the position that the violence that TNBGD people experience is gender-based violence and should be considered as such and accounted for in current thinking around gender-based violence. See further discussion on this in section, “Online gender-based violence: Towards an inclusive framework”.

The Left Out Project, in seeking to explore and understand TNBGD people’s experiences of OGBV, adopted a qualitative feminist internet research approach primarily influenced by standpoint theory, intersectionality, reflexivity and feminist ethics of care. 22 Interviews were conducted with participants to explore their experiences of OGBV. This data was then analysed using thematic analysis.

Qualitative research makes it possible to gather in-depth and meaningful knowledge that enables exploration of the experiences of TNBGD people with regards to OGBV.

Feminist internet research draws its influence from feminism, which seeks “to end sexism, sexual exploitation, and sexual oppression.” Feminist research interrogates and disrupts power, and has at its core the goal of dismantling systemic oppression to create a more just and inclusive reality. The FIRN methodological and ethical framework presents four key pillars to doing feminist internet research, namely standpoint theory, intersectionality, reflexivity, and feminist ethics of care. The pillars are discussed very briefly below. A more in-depth discussion can be found in the FIRN meta-research project report.

Standpoint theory places emphasis on the lived experiences of those who are marginalised and holds that “knowledge is always socially situated.” Standpoint is not only about knowing the conditions of oppressed groups, but about critically engaging with these positions, eliciting key insights, and using this knowledge and understanding for the purposes of emancipation and knowledge building.

Standpoint theory informed our research through making it clear to us as researchers that our participants have a wealth of knowledge from their lived experiences, and that they are better informed about their experiences than we as researchers could ever be. Adopting such an approach ensures that as researchers, we limit how much meaning we impose or assume about the research area and are primarily guided by the participants, allowing for a more careful understanding of the perspectives of TNBGD people in Botswana, Rwanda, South Africa and Uganda.

Intersectionality is the understanding that all forms of struggle and oppression are linked and overlap through race, gender identities, nationality, religion and sexuality. Crenshaw introduced the concept of intersectionality to show how discrimination based on gender and race, and other identity-based discriminations, exacerbate each other, and that they cannot be separated out. This informs us that our participants may experience multiple forms of discrimination that may come to be compounded and result in a unique experience of marginalisation.

Space was created for our participants to share how different aspects of their identity may overlap and intersect with the discrimination and oppression they experience as TNBGD people. The key identity aspect that was identified as intersecting with their gender identity is their sexual identity as people who identify as part of the LGBTQIA+ community. This is, unfortunately, not a surprising overlap and exacerbation of discrimination given the current climate of push-back on LGBTQIA+ rights and increase in hate-based anti-LGBTQIA+ rhetoric.

Reflexivity makes it possible for researchers to engage with their own positionality, power and privilege. 31 It asks that researchers critically consider their positionality, and how this impacts on the research process and their participants. Positionality allows researchers to engage with how their social location, privilege, values and assumptions, to name a few, may influence the research process. For instance, one researcher identifies as trans non-binary while another identifies as cisgender, and these positionalities, in addition to other identity experiences, shaped their experiences and interactions with the participants and each other.

Nyx McLean is the Southern Africa researcher on this project, based in both the Eastern Cape and Western Cape of South Africa. They are a white transgender non-binary queer person who uses the pronouns they/them, and have been involved in LGBTQIA+ advocacy and research since 2005. They shared how during this project, given the shift in global politics around LGBTQIA+ rights, especially transgender rights, they had started to notice an increase in harassment and hate speech. They experienced this personally in their offline spaces, and while this was distressing and difficult, the feminist ethics of care guiding the project ensured that we were able to speak this through and to also relate more carefully with the participants of our project.

Thurlo Cicero is the East African researcher and is based in Kigali, Rwanda. He works in international development with a focus on advancing access to information and technology to underprivileged, marginalised communities and those left behind. Thurlo is a mixed-race South African from Cape Town who is queer and uses he/him pronouns. He has been involved in LGBTQIA+ advocacy since he was 17 years old and volunteered at Triangle Project during university. He also formed part of the Cape Town Pride organising committee in 2004 and 2005 – advocating for the greater inclusion of Black and Brown causes within the community to highlight the critical issue of corrective rape, which targets our lesbian sisters in townships. The Left Out Project further expands his reach, using technology, to see how this research can offer ways to keep safe online while also advocating for the expansion of the definition of OGBV which will ultimately lend itself to better legal protections for TNBGD people within Africa and beyond.

Feminist ethics of care is centred on the concern for the research community and participants, and asks researchers to consider whether their work benefits participants and communities or is extractive and perpetuates injustice. Incorporating ethics of care into the research allowed researchers to consider the sensitivities surrounding OGBV and the violent experiences participants may have encountered. Using feminist ethics of care, it was possible to create an environment of care that responds to the needs of participants.

Researchers contacted LGBTQIA+ activists, organisations, and networks to enlist their help in reaching participants and to ensure that support structures for participants were in place when needed. It does need to be noted that given the state of LGBTQIA+ rights on the African continent, this is a particularly vulnerable group and their safety is of the utmost importance.

To facilitate this process of keeping the identities of the participants safe, especially in areas where TNBGD individuals are not protected, researchers opted to use aliases and asked participants to not utilise the video-sharing tool built into the video software and to instead make use of the audio function. Before recording, researchers discussed the research with the participants, as well as any possible risks to them, as a means of ensuring informed consent. Once participants had provided their consent, recording was initiated. The audio file was recorded directly to the researchers’ hard drives to avoid copies residing on third-party servers. As another measure to ensure that the information of participants is kept safe, the research team used Riseup for email communication.

Researchers worked to create an environment that allowed participants to be more comfortable and at ease. This included asking participants how comfortable they were, and how the interview could be adapted to make the online call more considerate of their needs. Participants were also told that they could freely opt out of the interview or retract their consent at any time.

Being fully aware of the legal ramifications this may have on some of the participants, researchers reiterated how participant information would be handled, processed and stored. Participants were also informed that their identities would remain anonymous. Virtual private networks were used to hide the location of the researchers to ensure their safety was also being cared for.

Sampling: The research was conducted in September and October 2022 with 29 adult individuals 18 years and older who reside in Botswana, Rwanda, South Africa and Uganda. Researchers selected two countries where LGBTQIA+ rights are legalised, namely Botswana and South Africa, and two countries where LGBTQIA+ rights are not protected, these countries being Rwanda and Uganda. The reasoning behind this would be to explore how TNBGD people’s experiences of OGBV would differ or if common themes would emerge across some or all of the countries.

The Left Out Project researchers divided the four countries into two regions, Southern Africa and East Africa. The reasoning for this decision was that the two researchers were stationed in South Africa and Rwanda, respectively. Due to the political environment of the countries chosen, the researchers had to determine the safest method and course to take in reaching out to participants. It became apparent that social media in Southern Africa was better utilised to communicate and garner interest, and so the researcher in Southern Africa opted to use social media in tandem with contacting LGBTQIA+ organisations directly for participation. On the other hand, in East Africa, freedom of speech is more curtailed on social media and tracking and tracing in this part of the world could have consequences for participants willing to engage; as a result, the researcher opted to only contact known organisations which are associated with minority, LGBTQIA+ and gender rights advocacy.

On naming, given the vulnerability of TNBGD people, we opted to use identifiers that are removed from their identities, including pseudonyms. This is because we are concerned about the possibility of our participants being identifiable. For instance, the current Ugandan context is especially risky for our Ugandan participants, and we are concerned for their safety. Instead, we opted to use letter and numbering identifiers. For Botswana, the letter B, Rwanda, the letter R, South Africa, the letter S, and Uganda, the letter U. The numbering assigned to participants is random.

Data collection: The study makes use of in-depth qualitative interviews, which were conducted online. Online interviews were conducted on a one-on-one basis through remote video recording technologies. Connectivity was not always good in all of the countries, and where connectivity failed outright on a first and a second attempt to have an interview, researchers would send the remaining questions to participants via email so that they could respond via text. The data was transcribed verbatim from the audio recorded, and interview recordings and transcripts were stored on our local devices and in a secure shared folder provided by APC.

Data analysis: The research team analysed the data using thematic analysis. Thematic analysis is a qualitative research method for analysing data collected from interviews and focus groups, as well as text-based content. Thematic analysis involves the identification, analysis and interpretation of patterns or themes that exist within the data. Steps followed in a thematic analysis include: 1) becoming familiar with the data, for instance, through reading over the transcripts several times; 2) coding or searching for themes – here we sought out what was common among all our participants’ experiences, as well as what was different or contrasting; 3) reviewing themes – here in our process we reviewed the themes identified in the previous step to determine whether they overlapped with other themes or were made up of further themes which could be separated out; 4) refining themes – here the decision is made on which themes to include in the final stage and why those decisions are made; and lastly 5) writing up the findings, which is where researchers draw the themes together and analyse and discuss them. Once these steps have been followed, the themes and related data are brought into conversation with literature to deepen the analysis. It bears noting that the themes that emerged from the data cut across all country contexts and so we have chosen to discuss these collectively and not separated out into country contexts. Where there is a difference in context, it will be noted.

In the above section we have discussed our conceptual framing, methodological approach, sampling, participants, data collection and analysis tools. We now go on to present the following: an inclusive framework for OGBV; TNBGD experiences of OGBV; online-offline-online movement of OGBV; navigating, managing, and remaining safe online; and recommendations.

Online gender-based violence (OGBV) is defined as: Acts of gender-based violence that are committed, abetted or aggravated, in part or fully, by the use of information and communication technologies (ICTs), such as mobile phones, the internet, social media platforms, and email.

OGBV may manifest in a number of ways. Some of the acts that are considered to be OGBV include: infringement of privacy; surveillance and monitoring; damaging reputation and/or credibility; harassment (which may be accompanied by offline harassment); direct threats and/or violence; and targeted attacks to communities.

Current conceptualisations of GBV and also OGBV neglect and invisibilise the experiences of TNBGD people by oversimplifying GBV as meaning violence against women – specifically, cisgender women.

There is some research in the way of GBV as experienced by LGBTQIA+ people, showing that they are “disproportionately victimized by gender-based violence,” but this group is still largely “an understudied population.” Williams et al. further speak to how GBV experienced by “sexual minority women and trans people is rooted in structural and cultural oppression represented in poverty, racism, misogyny, homophobia, and transphobia.” They argue that it is critical that in order to prevent GBV, systemic transformation is needed of the systems and structures that make this violence possible.

Research has shown that it is cisgender men who are the primary perpetrators of GBV regardless of whether their victims are cisgender women, TNBGD people or other cisgender men. Masculinity is enacted as an exertion of power over other bodies in order to maintain the gender order. Brubaker argues that feminist research must “connect gender to power” and that “this connection is key to understanding the relationship between the social construction of gender and the use of gendered violence.” This understanding needs to be extended to cisgenderism, which sees gender as a binary of cisgender women and men, and does not recognise or include other gender identities such as TNBGD identities.

It is this cisgender lens when applied to current conceptualisations of GBV that renders TNBGD experiences invisible, because they are simply not considered in conceptualisations around gender. For instance, misgendering is a common means of harm and violence enacted against TNBGD people as a way of invalidating their gender identity, but current conceptualisations of OGBV do not account for or take this into consideration. In addition to not being considered, TNBGD people and their experiences are often outright rejected as not real or valid, and are “neglect[ed] … within research, policy, and practice.”

The consequence of this is that resources, support and interventions are not made available for TNBGD people experiencing OGBV. For instance, information materials may rely on binary gendered language and visuals which “can reinforce perceptions that such resources are reserved for cisgender (not transgender) women.” A way to address this would be to use more inclusive language and visuals that signal that TNBGD people are included and welcome.

If TNBGD identities and rights are not recognised in a country, crime statistics may be skewed. For example, violence against transwomen may not be documented appropriately and included in GBV crime statistics. Graaff explains in her work how often transgender women are misgendered as cisgender men when their cases are documented, and this is then carried through in the statistical analysis of crimes. This narrow definition of GBV and gender identity is deeply harmful, especially considering that “across all intersecting identities, trans women are reported to experience higher rates of violence than almost any other group.” This is a key group that requires social, legal and other protective measures but is being excluded because current GBV frameworks are too narrow and are exclusionary. However, it does require noting that the expansion of the GBV definition to be inclusive of transgender identities must be careful not to perpetuate another binary – the transgender binary of transmen and transwomen. Non-binary and gender-diverse people are often not included in conversations about transgender identities, and it is critical that they are included – especially considering that this group makes up a third of the overall TNBGD community.

When asked, TNBGD participants in our study felt that their experiences counted as OGBV. B4, a non-binary person from Botswana, explained that this is because it is “based not just on the definition, but also kind of like the way like the violence specifically attacks my gender identity, I consider that as gender-based violence.”

Meanwhile, S1, a non-binary trans femme person from South Africa, said: "Yes. If you just look at the semantic definition of gender-based violence, it is violence directed at you, because of your gender. People are being violent towards me because I’m transgender, and that is my gender identity. Therefore, it is online gender-based violence."

B2, a transmasculine non-binary person from Botswana explained, “If you misgender me […] that means you’re invalidating me, and for me, that’s gender-based violence.”

As our participants have stated, the violence they experience is directly linked to their gender identity. S5 explains, “because of the fact that we have this gender, and that’s the qualifying factor for the attack. [...] Not even subtle, it’s definitely gender-based violence.”

The understanding of the violence our participants experienced would lend itself to the definition considered OGBV. We recommend that current definitions and examples of OGBV be expanded to include the experiences as highlighted by our participants and those of TNBGD people more widely, so that support services, structures, strategies, policies and other interventions may be appropriately designed and implemented.

The Left Out Project aims to specifically listen to what participants consider to be OGBV and advocate and lobby for the broader definition to include their experiences, which we deem to address in this four-country study. We aim to move beyond the oversimplification of OGBV as meaning violence against women and to instead define OGBV as violence that is experienced as a direct result of one’s gender identity and gender expression. If we do not expand our definition of OGBV, TNBGD people are at risk of continued discrimination and violence and we, as researchers and people working in OGBV advocacy, come to misrepresent the social reality and become complicit in gender violence against TNBGD people.

In seeking to expand the current definition of OGBV to be inclusive of TNBGD experiences, we asked our participants to share with us, if they felt comfortable enough to, their experiences of OGBV. The nature of the violence they experience is consistent with GBV and OGBV if we take into account that the reason they have been targeted is because of their gender identity. Participants first spoke of the relentlessness of the OGBV they experience.

U3, a transwoman from Uganda, shared her experience, saying, “Every time I open most of my social media apps, you will find an insult. […] It’s a daily thing.”

R4, a transwoman from Rwanda, shared that for her experiences of OGBV, “there are many times that I cannot count them.” S5 commented, “I can’t remember a time when there wasn’t transphobia online.”

The above speaks to the continuous nature of the OGBV targeting TNBGD people and that TNBGD people must navigate this violence online on their social media accounts. Some participants also expressed concerns over the safety of the internet connection in their countries.

R1, a transgender person from Rwanda, said that in her experience, “the internet is not good; it is not safe.”

Other participants spoke of how unsafe social media platforms were. R2, a transgender person from Rwanda, said that for her, “there is no platform that is safe.”

Meanwhile, other participants raised concerns about government surveillance. R8, a transwoman from Rwanda, said that “you hear that the government can easily access your data.”

R10, a transgender person from Rwanda, said that she felt “they are watching me, they are looking at my moves on the internet, it is not safe.”

U3 shared that “sometimes people use your internet connection to track you. […] They are tracking us through it. […] When it comes to transgender people and internet, nothing is safe.”

These fears are not unfounded, and social media and apps have been used to track LGBTQIA+ people, such as in the case of Andrew Medhat from Egypt. Police had located him through Grindr and then imprisoned him for “public debauchery” for talking to a man on the app. The way in which digital technologies are used for surveillance means that vulnerable groups such as TNBGD people need to have “secure basic protections that the general population might take for granted.” TNBGD people are “often an afterthought when it comes to user privacy and regulations,” for instance. This then places TNBGD people at an added risk of violence, because the digital technologies they use have not accounted for their vulnerability, and how parties such as state governments may make use of TNBGD people’s data.

Examples of violence

Our study found violence directed towards TNBGD people to be consistent with the definition of OGBV. The violence experienced by our participants was found to be in four key forms: dismissal of gender identity; sharing of images without permission; hateful comments; and threats of violence and death. Wilfully invalidating someone else’s identity based on their gender identity has been found to be linked to physical violence and also to have devastating effects on TNBGD individuals’ social integration and their mental health. The levels of self-harm recorded among TNBGD people are significantly higher when compared to cisgendered people due to the levels of violence and hate they are subjected to.

Dismissal of gender identity

Participants spoke of having their identity and gender questioned and dismissed as illegitimate; this is inclusive of refusing use of correct pronouns and active and intentional misgendering. Some reported that others online had made a sport out of trying to figure out what they deemed to be their “true gender”.

S8, a transmasculine non-binary South African who uses they/he/she pronouns, shared how they had been targeted in a group where: [M]y gender was unclear. Because there’s no pictures, there’s no expression. They were trying to guess what my gender was. And they were badgering me, “What are you? What’s in your pants? What are you?” The whole group chat turned into this game of trying to figure out what’s in my pants.

S8 continues to explain what made this interaction violent: [W]hat they were doing is not respecting me as a person. They were not respecting the boundaries that I’ve set and saying that I prefer you use they/them pronouns. It just went on as a hunting game. And I think it was more about trying to humiliate me and make me feel bad about myself.

B2 shared how when they changed their pronouns online, they received invasive and transphobic responses: You’ll get people inboxing you, “So, what are you? What’s happening with you? Are you okay? What is this they/them thing? Have you transitioned? Have you done this?” Like, why is that any of your business?

B6, a trans masculine-presenting person from Botswana, shared his experience of how someone had “intentionally used my dead name on their social media platforms.”

Meanwhile, B4 shared that the main form of transphobia they experience is through comments online with “people claiming, ‘We know you. So, this is what we know…’ And also, them telling other people on Facebook, ‘Don’t refer to them as that.’”

What B4 is describing here is how people lay claim to knowing what their gender is (the gender assigned at birth) and taking it upon themselves to tell B4 that they “know” them as well as telling others how to refer to B4. This is an outright refusal to accept and respect B4’s gender identity. There is a sense here that TNBGD identities are viewed as illegitimate or “not real”, and not taken as seriously.

Sharing of images without permission

The non-consensual sharing of images or “image-based abuse” can involve the sharing of images that were initially created with the subject’s consent but were not intended to be shared with others or images that were created without the subject’s knowledge or consent. Image-based abuse is GBV that our study found was predominantly directed at our participants due to their gender identities. This type of violence creates significant and long-lasting emotional distress, damage to personal relationships, loss of employment, and may lead to physical harm.

Our participants shared how their photographs had been shared on social media without their permission. B1, a transwoman from Botswana, shared how: We had participated in a pageant and pictures from the pageant were circulating all over social media platforms. I found out that my cousin had downloaded the pictures and shared them with my family to taunt me about my identity and my participation in the pageant. […] The pictures from the pageant were accessible to the public with no consent forms handed to us as participants to notify us of the use of our pictures from the pageant.

R2 said that she had experienced non-consensual sharing of images and that this led to harassment online when giving an interview to journalists: “Some journalists […] took my pictures and they put them on Instagram. People shared it on other social media.”

Another participant, R10, said that “it was my queer friends. They took the photos and uploaded them [online],” and the photographs drew the attention of “many people in the community.”

The risk to the safety of R10 as a transgender person in Rwanda was not given any consideration by the person who took the photos and shared them – even though this person sharing was also a member of the LGBTQIA+ community. The posting of images without permission, including the case of R10, emerged mostly as a means of ridiculing TNBGD people. The consequences of this, not only as being something that in itself is explicitly harmful and cruel, is the risk of a “witch hunt”. This has been seen before in Uganda with LGBTQIA+ activist David Kota, who was murdered after his name had been published in a newspaper.

We recommend careful consideration and reporting mechanisms for TNBGD people, where they are able to report that their image and information have been shared without their permission and actions to correct this are then made available to them.

Hateful comments

Our participants shared how they had received numerous hateful comments on their posts online. This is considered to be hate speech, which is defined by the United Nations as: Any kind of communication in speech, writing or behaviour, that attacks or uses pejorative or discriminatory language with reference to a person or a group on the basis of who they are, in other words, based on their religion, ethnicity, nationality, race, colour, descent, gender or other identity factor.

Participants in our study reported and described how people specifically commented on their gender identity and gender expression when they posted images online. For instance, participant U4, a non-binary Ugandan, explains how for them, “the very first time I experienced some kind of abuse and harassment was when I posted my picture in a dress. I received 1,000 abuses and harassment so I removed it after some time.”

U2, a transwoman from Uganda, also experienced something similar: “People started commenting, ‘Why are you doing this? You’re going to hell. You’ll die. We shall kill…’ I felt so shocked because people can say whatever they want.”

TNBGD people use social media in affirming ways in order to celebrate their gender identity, and this may include taking and sharing photographs of them in clothing that matches their gender identity. It is this content that receives targeted abuse online. Online acts of violence can be reflective of conservative ideologies as well as cultural beliefs and norms around gender identity, sexual identity, and expression of both gender and sexual identity, even in countries where such expression is not restricted through censorship laws. For instance, in August 2022, a Rwandan singer was jailed after others posted an image of her dressed “inappropriately” to social media. The attention drawn to the post resulted in her being arrested for indecent dress. An event such as someone being jailed for expressing themselves through dress may result in self-censorship of other women and TNBGD people both online and offline, out of fear of similar actions being taken against them – this has a direct impact on freedom of expression – as well as acts of violence such as harassment online, which is consistent with what our participants have reported.

Threats of violence

TNBGD people are often the targets of threats of physical violence, which studies have shown have severe and harmful effects on their mental and physical health and their ability to live freely and safely. 76 Types of threats often associated with OGBV are online threats of physical violence and assault with the intent to cause bodily harm, which includes sexual abuse. Transwomen of colour are particularly more vulnerable. 77 Participants shared how they receive threats of violence, including threats of sexual and physical harm and death threats.

B3, a non-binary person from Botswana, frequently receives direct messages with threats of sexual violence from strangers. They are concerned about this because “with Facebook, this is easy for someone to really track me and find me and probably kill me.”

B3 explained to us how they had participated in a TNBGD trend, and a few hours after posting this online “my girlfriend called me, ‘Are you okay?’ Yes, I’m okay. Then after a few minutes, ‘Are you really okay?’”

B3’s content had gone viral and received national attention. This attention was not positive or affirming, but rather transphobic and violent. It included “pastors quoting the Bible, people quoting scientific theories [about gender],” as well as more severe threats, which told B3, “‘You need to be corrected’ – people saying they are willing to violate me.”

What B3 is referring to here is “corrective rape”, which is enacted by heterosexual men on lesbian women and gender-diverse people in order to address what they believe to be a divergence from gender roles. B3 described this as “painful, honestly, to tell you the truth, because even my mom was freaking out. She didn’t understand what non-binary meant, but she did understand that this is cyberbullying and my child is not coping.”

S1 shared how fae had also received similar threats: I occasionally get death threats from anonymous email accounts. Which is wild, “There’s something in my spam that doesn’t look like spam. Oh, it’s a death threat. Cool, cool, cool new email that I can block.”

S4, a non-binary South African, has received comments on their posts about trans rights. “You now start hearing, ‘Transwomen are not women. I hate them. I would kill them.’”

These threats of violence are extremely distressing to our participants and deeply concerning, especially given the graphic and gratuitous nature of the threats, including threats of sexual assault and murder. TNBGD people need better reporting mechanisms and responses from platforms to ensure their safety. The consequences of these threats of violence on the mental health of TNBGD people are severe.

Reflections

Participants spoke of the internet and social media platforms as being largely unsafe for TNBGD people, and how when they posted any content online – be it their pronouns, images, thoughts or experiences – they were almost immediately targeted by transphobic individuals. Participants spoke of the continuous nature of the violence they experienced, and how some of their experiences were with regards to consent and sharing of their content by others to harm them or direct harm their way. They provided examples of violence which were grouped in four key ways: dismissal of gender identity; sharing of images without permission; hateful comments; and threats of violence and death. They were targeted by transphobic individuals as a direct result of their gender identity and gender expression. This is consistent with the definition of GBV, and shows that TNBGD experiences need to be accounted for in conceptualisations of GBV at all levels, inclusive of advocacy and policy.

The online and offline are not always separate. In previous conceptualisations around internet studies, online realities were seen as separate, but this thinking has shifted significantly with the uptake of social media. While someone may be harassed or targeted online, it does not mean that the violence only exists in the digital space; instead, the perpetrator may choose to seek out the person they are victimising and may engage in violence offline, such as stalking and physical and/or sexual assault. The same can be said for offline instances of violence which then migrate online, whereby a perpetrator may choose to continue victimising someone. An understanding of how people move between digital and offline spaces needs to be incorporated into our understanding of GBV.

While little research exists on the relationship between online and offline stalking among LGBTQIA+ people, a stronger link has been established regarding straight, cisgender individuals. Cisgender straight men engaged in online stalking were found to be much more likely to engage in subsequent offline stalking and harassment of their women ex-partners.

Directionality can be difficult to assess for online and offline violence, especially as online and offline interactions become more integrated in people’s day-to-day lives. However, in many cases, online and offline violence are closely related, whether one occurs after the other or the two work in tandem as part of a strategy of abuse.

Participants were asked whether they experienced violence online that then migrated to offline spaces or violence that moved from offline to online. At the time of this study, there was no available research on the movement of violence from online to offline or offline to online as experienced by TNBGD people. This is an area we recommend for further research.

Online to offline

Participants were first asked about their experiences of violence moving from the online to the offline. The reason for this was that we wished to understand if the GBV that our participants were experiencing was contained in the digital space or if it migrated to the offline, whether through geolocation or other means of tracing someone.

Participants shared that they had experienced the move of violence from online to offline. B2 shared: Yes, that has happened, where you would have a conversation with someone on Facebook and then it moves and then eventually you meet somewhere and they continue being that perpetrator.

B2 continued: It’s traumatising, it goes from being an online encounter, and you are sort of protected by the internet because they’re not in front of you and suddenly, they’re in front of you. That becomes really, really dangerous in that space.

Here B2’s experience speaks to the distress of finding that an experience was not contained or limited to a digital event, but also made its way offline, where they are no longer protected by a screen from physical violence. Having a perpetrator become a physical presence is dangerous, and puts one in an even more vulnerable position, where the threats made online may be realised as physical acts.

One means of gaining access to TNBGD people is through the use of apps such as Grindr, a geolocation-based app. Participants from Rwanda said that they had experienced violence online that moved to the offline, and how this had primarily been enabled through Grindr.

R2 said, “I met with someone, we met on Grindr but when I saw him at his place, they did bad things to me. I can’t even say it because I don’t even want to remember that time.”

R10 shared that “there was a time where I met a guy […] after conversations online. We met in person and they harassed me and actually beat me.”

U2 explained how Grindr has been used to track transgender and other LGBTQIA+ people: “Straight people started using it so that they capture some people. Some people were losing their lives, some people were stolen, some people were beaten, some people were violated.”

Here we see the use of Grindr, an app with geolocation capabilities, being used to target TNBGD people. Several similar cases have also been reported regarding the use of Grindr and other dating apps in order to lure LGBTQIA+ individuals to enact violence against them. Sharing locations online can also enable a stalker to track a person’s daily movements and learn details about their life, including the location of their home, work, and other places they frequently visit.

The system of geolocation used by apps like Grindr can potentially expose users to danger, since flaws in the app software can be taken advantage of in order to find the precise location of users without their permission. Furthermore, other forms of offline violence have been recorded to occur after online encounters, such as “sextortion”, bribery, and blackmailing. Individuals who are not yet or unable to be out are especially vulnerable to this type of offline abuse, since they are less likely to report it in fear of outing themselves in the process. Stalking also came up in our conversations. Online harassment and stalking may not be limited to the digital spaces that the perpetrator is targeting their victim through; they may escalate this violence and seek to find the person offline. This brings with it serious safety concerns for TNBGD people.

S4 shared the following: I’ve been stalked. I’ve been stalked. […] That guy created 16 profiles. I’d block him on one profile and he’d create another one. I remember there was a point where he would pop up where I was.

As S4 explained, this person stalked them online and then would appear in spaces offline: Every time I blocked him, he’d create another profile. I remember there was a point when he came to my house at 01h30 and he stayed outside my house until 03h00. It was so scary. It has been one of the worst experiences of my life, because I remember having to send my friends pictures and say, “If anything was to ever happen to me – today, tomorrow, any time – just know this [is] the person responsible.”

S4 takes action to ensure that their stalker can be identified should something happen to them as a result of this person’s attention. What is concerning about what this participant has shared is that they know that it is possible that their stalker may choose to harm them and that they need to put measures in place to ensure that the perpetrator is identifiable. This is not only concerning, but also evidence of a criminal justice system that is not safe for LGBTQIA+ people to access and would most likely dismiss the experience of a TNBGD person, and so instead of reporting the violence, the participant instead ensures that their peers know who the person is.

U1, a transwoman from Uganda, shared her experience of being stalked: [Someone] was acting like he was trans yet he wasn’t and he tracked us – me and my friends, we were three. He tracked us and we ended up with him and his friends, and they used force. They wanted to rape us.

She shared that this person had used Twitter and: [H]e opened a lot of accounts, with different emails, with fake names, and changing names on Twitter. So, he called us to the house, the apartment there […] and he ended up calling some friends.

In the experiences shared above, we see how perpetrators migrate from online to the offline, no longer only threatening TNBGD people in their digital spaces but actively seeking them out in their physical spaces. While TNBGD people can restrict public access, delete their social media accounts, or change their phone numbers, among other tactics, these may not deter a perpetrator – as we see in U1’s experience of someone opening multiple accounts and email addresses. Platforms enabling blocking and reporting of accounts may be helpful, but further action is required to ensure that perpetrators are prevented from opening multiple accounts. This is because while technology may enable violence, it is the perpetrators’ intentions to inflict harm that need to be addressed, and this requires systemic and structural interventions.

Offline to online

I guess it is harassing. But at first, they started out curious and asking questions. And I feel like obviously because it's an online space, they probably felt more comfortable. Even though we could just talk in person, but they are online. And then they started getting a bit more angry with their responses. (S8)

This was the experience shared by one participant of how someone they knew from offline spaces would contact them online.

The movement of offline to online violence is well documented in the context of cyberbullying and cyber harassment. The absence of direct contact helps perpetrators of online violence disassociate with their actions and limit feelings of guilt associated with acting immorally. Furthermore, online platforms make violent actions easier to enact, giving perpetrators easier access to their victims. Literature on cyberbullying also suggests that the lack of clear bystanders may make cyberbullies feel more confident to use online rather than offline platforms to target their victims, along with the potential for anonymity online.

Some literature exists about the movement of offline to online violence in the context of stalking. It explains that in the case of online stalking, which is primarily associated with ex-intimate partners, perpetrators use information gathered from in-person interactions in order to stalk and harass their ex-partners online. Reyns and Fisher warn that while a strong correlation can be found between online and offline stalking, cyberstalking may simply be another tool for a stalker’s “repertoire of available pursuit methods” rather than the result of offline stalking.

“Not yet”: The anticipation of violence

While some participants had not experienced the movement of violence from offline to online, participants from Botswana and Rwanda, B1 and R7, responded with “not yet” when asked about this kind of movement of violence.

This anticipation is worrying. It is not a matter of “no, not at all,” but rather of “not yet,” implying that encountering this type of violence is inevitable. Anticipating violence, while less severe than experiencing that violence, can still lead to significant negative mental outcomes. Within research on anticipated stigma and discrimination, anticipation acted as a stressor. Repeated exposure to this stressor eventually manifested in high rates of depressive and anxious symptoms. Similarly, constant anticipation of violence translates into the constant stimulation of a person’s “fight or flight” response, whose repeated activation can lead to serious, deleterious effects to a person’s health.

Reflections

In the above discussion we noted that the primary movement of violence was from the online to the offline, where perpetrators sought to gain access to their victims. This, as discussed, brings with it an added threat to the safety of TNBGD people, as perpetrators seek to make good on their threats through physical violence. In particular, we note the use of Grindr – and likely other location-based apps – to identify TNBGD people and to track and target them offline.

Most participants had not experienced the movement of GBV from offline to online, but what was concerning or important to note was the anticipation or view of this as inevitable through the responses of “not yet” when asked whether they had experienced violence moving from the online to offline or the offline to online. Where participants had experienced the move from offline to online violence, it was from people already known to the participants who wished to extend their contact with them.

Participants were asked about how they navigated and managed instances of OGBV, and how they remain safe online. We also asked them about how platforms responded to their reports of OGBV. TNBGD individuals face a disproportionate amount of harassment, discrimination and violence online, and it is important to take steps to mitigate these risks and create a safer online environment. The following key themes emerged from this exploration: awareness of reporting tools; blocking; solidarity strategies; avoidance; use of closed groups; engaging with transphobes; and responses from platforms when reporting.

Some participants showed an awareness of reporting tools on social media platforms. For instance, B2 shared that their understanding of reporting mechanisms is “where you can report a comment or anything negative about certain things or anything that is not inclusive or speaks ill about trans or gender-diverse persons.”

U2 also shared that “you can report someone using those tools […] and that account can be blocked or you can report and they remove [the] post.”

However, others (B1, R5, R7 and R2) said that they were not aware of reporting tools or how to report individuals who harass them online. It is concerning that there appears to be an absence of awareness of tools or a lack of uptake of the tools that may protect TNBGD people from OGBV. Reporting tools can help users flag inappropriate content, cyberbullying, and other harmful online behaviours which are associated with OGBV.

Blocking

Participants spoke of “blocking and reporting” (S2) as a means of managing their safety online. S6, a non-binary South African, shared that they “would just delete the comments, or ignore the DMs or block them.” U4 shared, “When I see something that is really not right […] I report and block.”

Two participants shared examples of incidents that resulted in them choosing to block. S1 shared how on Twitter “someone set up a DM bot once that just DM’d porn to me. I didn't know you could set up bots for DMing. I just ended up reporting that and blocking it, and it got suspended.”

Blocking came up as a response from an individual to another individual online, but it also emerged as something far more targeted that we refer to as a solidarity strategy.

Solidarity strategies

A solidarity strategy is a response to OGBV being experienced by another TNBGD person online. Mass blocking and reporting online refers to the act of blocking and reporting multiple users or pieces of content on an online platform. This can is done by marginalised people online for various reasons, such as to protect oneself from harassment and online violence, to report spam or abusive content, or to enforce community guidelines.

A non-binary participant from South Africa, S5, shared how when they see someone else being targeted: I report immediately, I don't question the degree of the transphobia. As soon as I feel like a trans person on the internet is in a vulnerable space and trapped in that engagement, I'll pull the plug on that immediately.

Meanwhile, one participant, B5, a transman from Botswana, explains how in his experience he no longer reads as transgender but rather “passes” and makes use of this passing strategically to intervene when other transgender people are being targeted. “Yes, then people listen,” B5 said. They listen because they assume he is a cisgender straight man who is calling them out on their transphobia.

S3, a non-binary person from South Africa, shared her experience of belonging to a group that actively blocks individuals who harass and troll TGBND people: Several of us will be reporting at once […] for example, that group […] if I want to report something, I put it on that platform. That platform has over 600 people and we are all going to actively report that account, so it will go down.

She continued: I don't navigate them at an individual level. So, if there's someone who is harassing, trolling, hate speech, [I take] screenshots and put [them] on the WhatsApp group, and then within a matter of like, 30 minutes, it's down. They've taken it down.

This is a strategic and coordinated response where one person is not burdened with trying to report an account alone. It also guarantees a response from the platform, which is then moved to block the user or remove the content.

Use of closed groups

Participants spoke of making use of closed groups as a means of keeping the TNBGD community safe. Closed groups is a function or option on platforms such as Facebook, as well as chat applications such as WhatsApp and Telegram.

R3 from Rwanda shared how she feels that these groups “are safe.” Participants attributed the reason for feeling safe to the fact that closed groups only accept people through an invitation process by people known to them. The use of closed groups is a tactic used by TNBGD people to keep each other safe. B2 explained: We found that if we have open groups, it allows for everybody to jump in and disrupt our home, what we're trying to create for ourselves. [We] make sure that we monitor and put security settings around it. […] We don't want to be engaging people on those platforms because they will derail the kind of work that we're trying to do.

Closed groups are considered to be safe spaces because of the support, advice, resources and community they offer members.

In addition to closed groups, participants spoke of other forms of keeping themselves safe. These include the management of the information they reveal about themselves. R2 shared that “I know how to use my social media; I don’t post anything when I know that it’s going to make me suffer,” adding, "I try to choose what I can post [and] I can hide some people [from viewing the post].”

Meanwhile, S8 manages their safety through privacy measures such as “not posting my pronouns. My safety measures would just be not talking about it or not making any inclination that I am genderqueer. Unless I know it's safe.”

Participants also shared how they made use of VPNs. U5 and R7 shared that VPNs give them a greater sense of anonymity and privacy. In addition to engaging in the abovementioned safety strategies such as blocking, use of closed groups, management of information and the use of VPNs, some participants spoke of engaging transphobes in conversation.

Engaging with transphobes

From our conversations on navigating, managing and remaining safe online, something interesting emerged. Some participants shared that they actively engage with transphobic users. R1 describes her reasoning for this as being that “I know that advocacy is a process because my purpose is to talk even if there are some people who might not change.”

B4 also takes this approach, sharing: If I talk to the person, I'm like, okay, this is how we do things. And really explain the situation to them. And then if they're like, “Oh my God, I didn't know that” and then apologise […] and then refrain from doing such things, then it is okay. I don't have to block them. But if the person just keeps hammering on then blocking is really usually my last resort.

Engaging with individuals as described by R1 and B4 is a strategy that requires patience and asks a lot of a TNBGD person in the way of emotional labour and exposure to potential violence. This is not an approach that everyone may have the capacity for – in particular with regards to mental health. We would like to propose that these acts of engaging with perpetrators and potential perpetrators of OGBV be considered to be “connective practices”, whereby TNBGD people attempt to connect and establish a link with antagonists in order to diffuse the situation and to, hopefully, shift their perspective to be more accepting of TNBGD people.

Responses from platforms

When asked about how the platform responds to them, participants reported that platforms responded in one of two ways to their reporting, either providing no feedback or outright dismissal of their experience.

B1 shared that in her experience, “as soon as I block the person, I never receive any other update regarding them or their presence.”

For their part, S8 and B2 also felt that their complaints via the feedback channels were ignored. Other participants shared how platforms outright dismissed their experiences or took action against them instead of the perpetrators. For instance, S1 shared that “on Twitter I've laid reports of what I thought was active harassment, but they thought it wasn't.”

Meanwhile S7, a non-binary South African, shared how someone had commented on a post of theirs with threats and the response from Facebook was “it was my post that was deleted” and “I got my account restricted.” When asked why they think this happened, they shared that “if there are people moderating, there is internalised misogyny there that is so deep, that they're like, ‘Yeah, this is okay.’”

One participant, S6, shared that in their case, they found that language was critical in whether an account would be blocked or not, explaining: For the DMs, especially in languages that are not English, it's really harder to get somebody's account taken down or understand insults and slurs in the language. I can report and block but I don't think reporting will do anything.

This shows a platform approach to reporting that does not account for language and the ways in which OGBV manifests in languages outside of the platform’s primary language – usually English. For instance, in 2019 it was reported that out of 111 different languages officially supported by Facebook, the community standards of the platform had only been translated into 41. This is only in the way of community standards, which places the expectation on the user to understand that content such as hate speech and threatening violence are not accepted on the platform.

What is further needed is an understanding of the moderation processes, who moderators are, their regions, and the languages spoken, in order to understand the nature of the support available for users reporting instances of OGBV and other violence. There have been several calls for greater transparency in the moderation processes of platforms.

Reflections

Participants mostly used blocking as their primary tool to protect themselves, as well as managing their contacts, while some participants are in closed TNBGD groups which require an invitation or a vetting process before a member is allowed to join the group.

Some participants spoke of how they manage what information they share about themselves, such as their names and location, to prevent being traced offline. Some, concerningly, spoke of leaving platforms and/or silencing themselves because the OGBV had become too overwhelming. Most participants did not feel that their cases were being taken seriously or that OGBV was recognised as a serious matter that requires urgent addressing by platform owners.

An interesting and strategic response to the lack of response from platforms to individual reporting is the use of a solidarity strategy whereby a group of people all report an account individually in order to gain the platform’s attention. This coordinated effort shows care for the individual who may have been targeted as well as the collective in order to protect other TNBGD people from being targeted.

When reporting violence, participants were told by the platforms to block the accounts, and after reporting, most participants did not receive feedback from the platforms, so they were not aware if the platforms had dealt with the matter and whether other TNBGD people were being targeted. This leaves participants, and other TNBGD people who report OGBV, in a liminal space of not knowing how their cases have been managed or processed. Greater transparency is needed from platforms in how they managed cases, as well as open communication to ensure that they do not exacerbate an already difficult experience.

In this report we have presented the OGBV experiences of TNBGD people from Botswana, Rwanda, South Africa and Uganda. We have made the argument that the conceptualisation of OGBV needs to be redefined to include the experiences of TNBGD people in order to be more inclusive, responsive and ethical. If the conceptualisation of OGBV is inclusive of TNBGD experiences, responses to OGBV will be inclusive and better suited to the needs of TNBGD people.

The TNBGD community is not a fringe community or a minority that can be ignored when we think of OGBV, as they currently face a disproportionate amount of violence online and offline – as Graaff has shown, transwomen in particular experience the highest rates of violence. 111 It is urgent that the necessary steps are taken to understand and respond to TNBGD people’s experiences of OGBV.

The Left Out Project makes the following recommendations:

TNBGD individuals can take the following measures towards increasing their safety by:

• Using privacy settings on their social media accounts to control who can see their posts and personal information.
• Being selective about whom they share personal information with, such as their location, phone number or email address, and not sharing it with people they do not know.
• Utilising the block or report feature on the platform that they are using to prevent perpetrators from contacting them again.
• Considering using a pseudonym instead of their name online to protect their identity.

The LGBTQIA+ community and allies can help fight OGBV by:

• Calling out and reporting hateful and discriminatory comments and behaviours online and offline.
• Educating others about the harmful impacts of transphobia.
• Showing support for TNBGD individuals by using inclusive language and respecting the pronouns and gender identity of TNBGD people.
• Advocating for policy changes that protect the rights of TNBGD individuals online and offline.
• Helping to amplify the voices of TNBGD individuals by consensually sharing their stories and experiences and supporting trans-led organisations.

LGBTQIA+ organisations can support TNBGD people by:

• Being more inclusive of TNBGD people in their organising, language choice and treatment of TNBGD people.
• Providing resources and information on online safety that includes tips on how to protect personal information, how to report harassment and where to find mental health support.
• Offering support services such as counselling and crisis hotlines for TNBGD persons who are experiencing OGBV.
• Providing training and education for staff and volunteers on how to recognise and respond to OGBV, and how to be respectful of pronouns, identities and experiences.
• Helping create online safe spaces such as moderated forums and support groups for TNBGD individuals to connect with each other and find community.

Organisations addressing GBV/OGBV can support TNBGD people by:

• Adopting an inclusive understanding of gender, gender identities, gender expression, and OGBV.
• Adopting an intersectional approach to their work that is inclusive of TNBGD people and how they may face compounded forms of discrimination and harassment based on their gender identity, race, ethnicity, disability and other factors.
• Offering support services such as counselling, health services and crisis hotlines for TNBGD individuals who are experiencing OGBV.
• Advocating for media representation and inclusion that are inclusive of TNBGD individuals to help reduce negative stereotypes and promote acceptance
• and understanding.
• Legally recognising TNBGD identities to pave the way for future laws protecting the well-being and safety of TNBGD people, including in the context of OGBV.
• Including TNBDG sensitisation in school curricula, police training, etc.

Civil society and development organisations can:

• Advocate for policy changes that protect the rights of TNBGD individuals online by seeking to prevent online hate speech and discrimination.
• Raise awareness about the prevalence and impact of OGBV on TNBGD individuals through public campaigns.
• Provide funding for research that focuses on online safety for TNBGD individuals.
• Train all staff and volunteers on how to recognise and respond to OGBV in an inclusive and respectful manner.
• Provide grants to secure VPNs for LGBTQIA+ community members in countries with discriminatory governments and criminal justice systems.
• Forbid the non-consensual sharing of photos, videos and information of TNBDG individuals in organisational policy.

Social media and digital tech platforms must:

• Engage with TNBGD individuals and communities to understand the struggles they face using online platforms and design better, more inclusive platforms.
• Develop clear and comprehensive policies that prohibit online harassment, bullying and hate speech that contribute towards OGBV and ensure that they are enforced consistently and with care.
• Provide inclusive and respectful reporting tools in different languages that allow users to report OGBV.
• Create educational materials about blocking, flagging and reporting perpetrators of OGBV (e.g. in the style of a “tour” of new features).
• Protect the privacy and personal information of TNBGD individuals and ensure that their identities are not revealed without their consent.
• Use ethical AI to identify and remove harmful content and prevent the spread of hate speech and harassment of TNBGD persons online; this should also be inclusive of slurs and hate speech in other languages.
• Include mechanisms to confirm the identities of users and include “verification” tags in order to safeguard platform users.
• Commit to never sell user data to any third party and to never share data with governments or criminal justice systems.