Power X Expression X Violence: Women’s freedom of expression on social media in Malaysia

The research aims to address two dangerous assumptions: first that access to, and exercise of freedom of opinion and expression, is equal for all; and second that social media platforms are inherently emancipatory. Given how, historically, neutrality and egalitarian values in effect privilege cisgender men and cisgender men’s experiences, such assumptions obscure rather than enable a diversity of voices and inevitably trivialise the cost of online GBV.

In 2015, a Facebook post went viral: it described how the author planned to break into the house of a 69-year-old woman, Datuk Noor Farida Ariffin, and sexually assault her after she had called for a review of the Sharia laws, including those related to khalwat.[2] She is the spokesperson for G25, a group of former high-ranking civil servants that encourages rational and progressive discourse on Islam. When told off (on Facebook) by someone else who said that was not funny to joke about sexual assault, the author of the post scoffed and retorted that he was exercising his right to freedom of expression.[3]

In a separate incident, a woman was harassed on multiple social media platforms after she called out sexism in an article that conflated the purchase of a pair of cufflinks with fellatio as a great gift to one’s boyfriend or husband for Christmas. Among others, the attacks denied the article as sexist and claimed the woman was unable to take a joke. The demand for an apology and retraction of the article by feminist activists was deemed a form of censorship by human rights lawyers and others, and therefore self-defeating to the principle of freedom of expression.[4]

These two instances point squarely to the lack of understanding of freedom of opinion and expression. It is apparent that the demand for an absolutisation of freedom of opinion and expression and call for a blanket rejection of any form of censorship risks silencing and punishing women. Such an approach assumes that men and women enjoy equal access to and exercise of this freedom. It also overlooks sexism, discriminatory remarks, and online GBV and passes these forms of speech as speech protected under this right. When harmful speech is weaponised against women’s freedom of opinion and expression, it is weaponised against their public and political participation. Although social media has reduced the barriers to women expressing their opinions and thoughts, access to social media does not necessarily translate into this freedom being equally enjoyed by women and gender non-conforming persons, given the historical and structural inequalities that are also reproduced on technological platforms.

It is apparent that the demand for an absolutisation of freedom of opinion and expression and call for a blanket rejection of any form of censorship risks silencing and punishing women. Such an approach assumes that men and women enjoy equal access to and exercise of this freedom.

In Malaysia, and to some extent, globally, gender inequality is often and rightly addressed in terms of GBV and gender discriminatory impacts. However, the impact of gender inequality in relation to freedom of opinion and expression is largely unaddressed. A framework for an unrestrained freedom of opinion and expression means very little to women if it ignores the inherent unequal power dynamics in our access to human rights and equal protection under the law.

This research seeks to develop substantive evidence that could contribute to the development and refinement of arguments for women’s equal access to freedom of expression on social media. It recognises how these online spaces can effectively restrict and limit women’s public and political participation, and as a result, deny women the right to shape and reshape the dominant narrative. More importantly, the research calls for attention to the intersections of gender, sexuality, ethnicity and other social locations producing multiple standpoints. The research, therefore, focuses on unearthing the power dynamics of various forms of expressions and the intersecting identities of women; how our current understanding and practice of freedom of expression on social media have allowed online GBV to grow with impunity, normalising extremism and GBV, and; how the freedom of expression is asserted by women and what the subsequent responses to it are.

In selecting the research methods, the team made a conscious decision to position lived realities of women, in all their diversity, at the centre of knowledge-building. The intention was to unearth the power dynamics of various forms of expression from the women’s standpoints. Like other research that adopts feminist methodologies, this research challenges the notion that knowledge is only situated in the researcher. Deliberate efforts were therefore made to draw out the voices and experiences of women that are often invalidated and overlooked because they do not fit the mainstream perspectives.

Interview as a qualitative research method was selected as the main method to bring forward the narratives and experiences of 23 women (including lesbian, bisexual, transwomen and gender non-binary), complemented by desk research and a two-day participatory workshop. Semi-structured interviews with open-ended questions were used for the interview process so that the women were better able to enter into the research process as active agents and not merely as objects of scrutiny. A snowball sampling technique was used to identify participants who are not within the immediate network of civil society organisations. The process started with identifying few women from the community of queer persons, young women studying in university and women who identified as feminists on social media and are not affiliated with civil society organisations. These women were then asked to identified other women who had also experienced online gender-based violence in the last few years.

Two main categories of research participants were identified for this research: women and lesbian, bisexual, transwomen and queer (LBTQ). As the researcher, I recognise that the separation between women and LBTQ is a misleading construct and implies that the LBTQ are not women. It is, however, necessary to explicitly articulate the broader structural dynamics that tend to invalidate and make invisible the narratives of LBTQ women, even within the feminist and women’s rights movement. Three women identified themselves as non-binary but acknowledged that they present as women to the world.

Deliberate efforts were therefore made to draw out the voices and experiences of women that are often invalidated and overlooked because they do not fit the mainstream perspectives.

Five aggressors were also interviewed to better understand their interpretation of their behaviour and motivation for behaving aggressively towards women. Approaching the aggressors was not easy as the identified aggressors were not within the researcher’s immediate network and they predominantly conversed in the colloquial Malay language. To work around the lack of access, the researcher started with creating an anonymous Twitter account that follows the accounts perceived to be from the conservative group, through which the researcher discovered several accounts that are from the conservative network. About a total of 30 interview requests had been shared, but only five responded favourably. Two others had indicated interest but decided to pull out at the eleventh hour. In two interviews (one online text-based interview and one face-to-face interview), the researcher had to anonymise herself, partly for security, and partly to avoid the disposition of the researcher as a liberal, feminist and LGBT rights advocate. In addition, the researcher reached out to a list of aggressors through their Twitter.

These interviews were not meant to triangulate women’s experience of online GBV. For this research, an aggressor is defined as someone who has been part of an online GBV incident by making a deliberate expression on social media that caused violence or led to the circulation of violence against others. While there were only a few interviews, these were conducted to inform the analysis of the data, and to better understand motivations behind the aggression, recognising fully that deeper analysis of online GBV is much needed.

As mentioned, a two-day participatory workshop was convened with eight women, seven of whom were interviewed for the research. One woman who had declined the research interview agreed to attend the workshop. During the workshop, preliminary findings of the research were shared with the women. This workshop was extremely important to check biases and misinterpretations of the data. It proved incredibly useful as, while the women shared their experiences, discussed and reflected, they also discovered new insights for themselves.

The decision to keep everyone anonymous, including the aggressors, was taken to prevent further visibility of those who had experienced violence and the possibility of further violence. Relevant information that may lead to the identification of the women was also anonymised. However, the virality and high visibility of some of the cases meant that full anonymity is not possible. Given the circumstances, the organisation made a decision to limit the circulation of this research to trusted allies. Findings of the research will be disseminated through this research summary and infographics.

The inevitable risk of online GBV means women have to navigate between opportunities and vulnerabilities through the mediated performance of the self digitally.

The findings of the research are categorised into three main parts. The first, titled “Expression of the (digital) self” looks at how women express themselves and perform on Facebook, Twitter and Instagram.

While social media has accorded women opportunities for more visibility and spaces for expression, it does not mean that women are finally liberated and free to engage publicly and politically. The realities of the women in the research show that their expression and performance of self are still subject to the alternate technological mode of society which takes on the unequal power dynamics of existing social and gender norms.

All women in the research, despite their various backgrounds, spoke about online GBV as inevitable to their existence on social media and that the “mind does not register it [as violence] anymore”. They commonly expressed a shared sense of fear and vulnerability on social media. To exist is to risk being harmed. The inevitable risk of online GBV means women have to navigate between opportunities and vulnerabilities through the mediated performance of the self digitally.

Even though there have been some (not many) advances in women’s public and political participation, their visibility remains highly regulated by cultural values and societal expectations. A media professional was unable to post content about her personal fitness as it reduced her to a sexual object and the content was deemed “unsuitable” for her professional image as a current affair journalist. In this sense, the decisions the women made in constructing their digital selves and bodies (through the choice of names, photographs, profile descriptions, friends or followers lists and the type of content they made or shared) were guided and conditioned by their existing experiences and societal labels as women, Muslim, lesbian, fat, liberal, feminist or other social identities.

Even though there have been some (not many) advances in women’s public and political participation, their visibility remains highly regulated by cultural values and societal expectations.

One main strategy used by women to navigate power digitally was through the fragmentation of the self across different social media sites, based on the techno-social design of each platform. Instagram, as an image-focused platform, is particularly powerful for self-expression through selfies, image-curating and self-branding. It is a site treated by most of the women as a non-political platform [5] or a space for trivial narratives and for some, it is used to share “happiness” and “the best version of self”.

Instagram facilitates a somewhat compulsory performance of the aspirational version of ourselves. These standards are predominantly dictated through the algorithm and the lens of the audience, which takes on the values of a heteronormative society. This in itself is a barrier to many women expressing a part of themselves that does not conform to gender norms. The pressure to “fit in” according to these standards means that a Muslim lesbian is unable to share her happy moments with her partner or a trans fem person is fearful to post photos of them wearing make-up.

Facebook and Twitter are the two platforms most women used to express their political thoughts and are often compared with each other. Facebook, compared with Instagram and Twitter, is the platform that carries most women’s embodied experiences and identities, and it is also there that “friends” are known personally and could include family members. And it is because of the lack of anonymity on Facebook that there is a barrier to women’s equal access to (and exercise of) freedom of opinion and expression. The ability to anonymise in full or in part, in effect, ensures access that may have been denied to women because of gender norms, gender stereotypes and unequal gender-power dynamics. To some extent, anonymity is essential to the protection of one’s physical safety. A Muslim woman in her mid-twenties chooses to stay semi-anonymous on Twitter after she was trolled and threatened with physical violence following her political expression several times. While remaining visibly as a Muslim woman, she uses an alias and never reveal too much of her personal details on Twitter, while stay unequivocal in expressing her thoughts around social inequalities. Anonymity promotes freedom of expression of topics and narratives that are normally censored and erased by the status quo.

The same tool of anonymity used by women to access freedom of opinion and expression is also exploited by aggressors in order to escape accountability. One aggressor shared that he is more aggressive on his anonymised Twitter account and would say things that he knew his friends and family would disapprov of. Having said that, using their real names did not stop the three other aggressors from perpetrating aggression online. In fact, one of the aggressors felt that the generic nature of his name amidst a vast social network accorded him some level of protection for his aggression on Facebook and Twitter because he knew it will be almost impossible for others to trace it back to him. However, after seeing how everyday people can be doxxed and attacked makes him think twice before posting any comments online. Essentially, it is the sense of impunity and knowing that they can get away with the abuse that underlies the violence, and anonymity is a means through which they can achieve that.

The same tool of anonymity used by women to access freedom of opinion and expression is also exploited by aggressors in order to escape accountability.

In addition, our ability to freely express ourselves must be located within the context of surveillance through spectatorship and interactions with others on social media. Facebook is a space where most women’s social context collapses – it is an accumulation of family members, friends and acquaintances from their physical lives who may or may not share the same values and interests. Whereas on Twitter, women are better able to network and connect with strangers who share the same goals, attitudes, and values as them. It is through our tweets and narratives that our identities are expressed and communicated. The ability to freely express ourselves or the need to censor our opinions are not isolated from cultural discourse, structure and practices within our networks.

The visible queer narratives on a queer Muslim woman’s Twitter account have allowed her to connect with like-minded people where she “found people [to be] much more aligned with [her] ideas”. One photograph of a cis gender woman would receive contrasting responses on Facebook and Twitter. On Facebook, her family members would comment rudely on her body size whereas on Twitter she would receive affirmation from her followers. She restricted herself from speaking about gender equality and LGBTQ issues on Facebook, knowing that her family members would not approve of it. The people and context around us can force us to make very deliberate and conscious decisions around our expression and self-censorship.

Women’s performance and expression of self are based on decisions that are never frivolous or casual. They are driven by embodied identities, social locations and structural inequalities and technological architectures and political visions of these social media platforms. Movements and efforts to ensure equal access and exercise of freedom of opinion and expression should start with acknowledging the inadequacy and the gaps in our knowledge when it comes to lived realities of those who are often ignored and marginalised. It means nothing should be created for us without us.

Women’s performance and expression of self are based on decisions that are never frivolous or casual. They are driven by embodied identities, social locations and structural inequalities and technological architectures and political visions of these social media platforms.

Part II: Expression and violence

The second part of the findings investigated the messy entanglement of online GBV with the intersectionality of identities and the inherent unequal access to freedom of opinion and expression; and how this entanglement of conflict and power manifests itself within a complex system of global social media companies, algorithms, national governments, social structures and people. While all the women experienced online GBV because of their identity as women, the nature, intensity and impact of the violence differed based on their various intersecting identities and social locations. The stories of young women, queer Muslim women, women with disabilities, women journalists and fat women in this research illustrate the different ways in which violence is perpetrated and experienced against one’s intersectional identity. Understanding how oppression and discrimination mutually construct each other to create a unique experience for each woman is key to generating appropriate context-specific responses when addressing online GBV. For instance, for Nadia, a woman with disability, the aggression against her #metoo story was compounded by social isolation and the stigmatisation of people with disability as being sexually undesirable, and by their perceived inability to be a functioning member of society. Addressing online GBV for women with disabilities would also require consciousness-raising around the rights, dignity and autonomy of people with disabilities.

While all the women experienced online GBV because of their identity as women, the nature, intensity and impact of the violence differed based on their various intersecting identities and social locations.

The women’s stories also show that the violence they experience includes normalised abusive acts that are harmful in aggregate but do not meet the legal threshold of hate speech or criminal behaviour. This alludes to the broader and stickier conversation on the power of language and discourse. In one instance, the aggressor tweeted about the presence of LGBTQ people at the 2019 women’s march in Kuala Lumpur. She describes herself as an advocate for the causes of the women’s march but believes being LGBTQ is against her religion and that the march was hijacked by the LGBTQ communities. In her tweet, she invited people to comment on the presence of pride flags in public spaces and insinuated that “this is where Malaysia is heading now”. Her tweet was retweeted and shared by thousands of users and many responded with abusive and violent comments against LGBTQ persons, including death threats and hate speech. The absence of violent language in her tweet is justifiable to her as a legitimate expression of her views and political stance, even though it cascaded into hate and aggression by others towards LGBTQ people.

Our language and discourse are not neutral and they engage with various structures and institutions of power to regulate our behaviours and expressions. Language and discourse produce and reproduce meanings, norms, stereotypes, otherness and discrimination. The absence of direct violent and abusive language in an expression does not necessarily remove one’s responsibility regarding the perpetuation of harm, especially when they are in a position of power and influence to incite people to react negatively to the issue or person they had targeted.

The absence of direct violent and abusive language in an expression does not necessarily remove one’s responsibility regarding the perpetuation of harm.

These sorts of expressions often fell through the cracks of “community standards” of Facebook, Twitter and Instagram and in effect, disguised online GBV as “freedom of expression”. Those who are targeted face an increased risk of further violence with every single signal boost. At the height of the online mob attacks against the 2019 women’s march in Kuala Lumpur, the hashtag #womensmarchmy was also used by curious bystanders and people seeking business opportunities on Twitter. Even though they had no intention of causing harm, the algorithm was unable to discern the quality of every interaction and worked towards amplifying the harmful content that had the most interactions. The economic incentive of the algorithm to maximise the distribution of trending contents, which are usually boosted by online GBV is counterintuitive to policies against hate speech and GBV claimed by social media platforms. The harm is heightened when our appreciation of freedom of opinion and expression is void of the gender lens, and narratives and discourse perpetrating online GBV are seen merely as an exercise of one’s freedom of opinion and expression.

Part III: Responding to violence

The third part of the findings looks closely at the strategies and responses adopted by the women in the absence of effective remedies to online GBV and the violation of women’s freedom of opinion and expression. Given the failure of institutional support, much of the effort of responding to violence is transferred to the individual women and, at times, the collective response of feminist networks or women’s rights organisations. Only two out of twenty-three women interviewed made or attempted to make a police report about the online GBV they had experienced. Even then, law enforcers victim-blamed them and indicated that the harm was not serious. Given the circumstances, the “block” and “unfriend” buttons were regularly used by the women as part of their coping mechanisms. Some had even adopted the tactic of preemptive muting by blocking trolls who had harassed other women. Many of the women had realised that people were not there to have a conversation and it was counterproductive to engage. Blocking, disengaging and unfriending were important tactics for many women involved in this research, to avoid exposing themselves to hateful narratives and attacks.

Given the failure of institutional support, much of the effort of responding to violence is transferred to the individual women and, at times, the collective response of feminist networks or women’s rights organisations.

Naming-and-shaming is another strategy that some of the women employed. One screenshot and tweeted all the misogynist comments against her in a long thread. In the interview, she shared it as a powerful moment in which she was able to reclaim her narrative. Many sexual harassment survivors had similarly called out their perpetrators on social media, following the failure of institutions or law enforcers to address their issues. The culture of online vigilantism is reflective of the fourth wave feminist movement: individuated, micropolitical and do-it-yourself action.[6] It has proven to be effective in creating awareness and public discourse. However, often the individual bears the risks and costs i.e. increased physical vulnerability or lack of physical security, isolation, alienation, mental health issues, potential defamation suits, and the lack of employment security.

The culture of online vigilantism is reflective of the fourth wave feminist movement: individuated, micropolitical and do-it-yourself action.

However, women refused to sit and wait for the authorities to fix the issues they face and are bypassing the institutional barriers to justice by doing it themselves. In some cases, the act of online vigilantism raised questions of boundaries and ethical considerations. Aggressors in less powerful positions received harassment and violence too. While the act of speaking up is powerful and even necessary to break the status quo, previous instances have shown that naming and shaming may be confusing for bystanders as most people are fixated in pinpointing one victim versus one aggressor, even though both could have experienced online violence. Also, these actions do not address the fundamental structural barriers for survivors to access justice and the cultural and societal prevalence of sexism that has worked against women survivors.

Moving forward

Social media as a space for expression is highly contested as gender norms are persistent and yet, women are constantly pushing back and disrupting the normative discourse, at the risk of experiencing aggression and violence. More than anything, countering online GBV does not stand in contradiction with freedom of opinion and expression. It requires an expansion of access to that freedom so that women and vulnerable groups can express themselves without the risk of reprisal from state and non-state actors.

Addressing online GBV means holding people accountable for their individual power and privilege and the manner in which they exercise their freedom of opinion and expression. It means to be actively aware that we live in a world where systemic discrimination against women is normalised, and often reproduced in the digital and social media space. Therefore, to stay neutral is to perpetuate systemic discrimination and obstruct equal access to the right of freedom of opinion and expression. It requires us to understand that these freedoms are not equally accorded to everyone and online GBV is an important manifestation of unequal gender-power dynamics that dominate all spheres of our lives.

Policy alternatives

The elimination of online GBV requires a rethinking of current strategies on such violence as it involves a range of rights – freedom of expression, the right to political participation and rights to non-discrimination, dignity and safety, which are in turn perceived as competing rights to vested interests in the politics of the country. It also involves globally dispersed actors – government, citizens and digital platforms; the latter often located in a different jurisdiction.

KRYSS Network’s research has also shown that the design and infrastructure of social media can fuel and encourage hate. In particular, this can result from the underlying economic structure of platforms like Twitter, Facebook and Instagram that rely on interactions among people. The more interactions there are, the more data the platforms will capture, which then informs the algorithm. The algorithm is designed to maximise the spread of viral content and this often includes inflammatory content. Spaces like Twitter also restrict nuanced conversations and encourage debate and polarised arguments. At this stage, more research is needed to guide and experiment with different designs of platform architecture so that technologies can be employed to address the underlying issue of gender inequality.

While laws are necessary in cases where severe harm is caused, online GBV also comprises a range of abusive behaviours including targetted harassment, coordinated mob attacks, and deliberate stigmatisation and discrimination that does not amount to criminal harm, even though it may lead to an aggregation of harmfulness, especially in cases of an online mob attack.

There is a need for a multi-stakeholder approach and an inter-governmental agency response to online GBV. This would require the deployment of multifaceted strategies, from the macro to the micro level, involving laws and policy reforms, implementation and the expansion of prevention programmes, research and monitoring etc.

Policy recommendations for the state

• Conduct nationwide empirical research to better understand people’s experience with online GBV, including why people perpetrate hate and violence online. The collection of data is key in understanding where it happens, who suffers, who are the perpetrators and why it occurs. The findings could then inform the design of victim-centric policies and laws, as well as the use of alternative approaches to ensuring justice for victims, thereby increasing the effectiveness and relevance of such laws to the victims and survivors of online GBV.
• Employ good prevention strategies and victim-centric redress mechanisms in learning institutions, including schools, colleges and universities.
• Address internet etiquette and behaviour, as well as the importance of having non-discriminatory and respectful interactions online through educational programmes and suitable self-regulatory and monitoring mechanisms.
• Provide guidance on what to do if bystanders witness an incident of online GBV.
• Review the standard operating procedures of government agencies and the police in dealing with complaints of online GBV, to be completely victim-centric, timely and proactive.

• Conduct regular public campaigns on raising awareness on hate speech and online GBV.

• Conduct an evaluation and audit on existing responses by Malaysia Communications and Multimedia Commission, Malaysia Communications and Multimedia Content Code, Cyber Security Agency and the police in addressing online GBV and to provide recommendations for improvement, including establishing gender desks with the necessary expertise for response, and to add specific content and guidelines for cases of online GBV, detailing approaches to be adopted to address it, or more specific guidelines to address it in specific forms.

• Set up a response team for online GBV where content takedown is urgently needed. This is especially required in cases involving the non-consensual dissemination of intimate images and personal data.
• Enact the Gender Equality Act, the Sexual Harassment Act and amend the Penal Code to make stalking an offence.
• Expand Article 5 of the federal constitution to include explicit protection of the right to privacy.
• Establish an inter-ministerial task team and conduct regular multi-stakeholder consultations with the relevant ministries, government agencies, civil society, academics, media and social media platforms.

[1] The term “women” is used to include cisgender women, transgender women and female-presenting who identify as non-binary.

[2] Close proximity between persons of the opposite sex.

[3] Maysia Insider. (2015, 23 December). No ‘rape’ threat just sarcasm, says man of post against G25’s Noor Farida. The Edge Markets. https://www.theedgemarkets.com/article/no-‘rape’-threat-just-sarcasm-says-man-post-against-g25’s-noor-farida

[4] New, S.Y. (2014, 5 February). Why so Serious? #Fellatio. Loyar Burok. https://www.loyarburok.com/2014/02/05/serious-fellatio/

[5] This may have changed after the resurgence of #BlackLivesMatter movement following the death of George Floyd in May 2020. Since then there has been an increase of social justice content on Instagram.

[6] Jane, E. A. (2016). Online misogyny and feminist digilantism. Continuum, 30:3, 284-297.