White paper on feminist internet research

The APC WRP has been working towards imagining and making a feminist internet by building and strengthening networks of researchers, activists, academics, thinkers, coders, artists and others. As part of the Feminist Internet Research Network (FIRN) project, supported by the International Development Research Centre (IDRC), this white paper aims to assess feminist internet research in relation to internet governance and policy, with a particular focus on scholarship in the global South.

Out of all the topics identified in a mapping exercise (elaborated in the next section on methodology), the white paper explores in depth eight topics in order to understand a feminist approach to these subjects, key areas of analysis and boundary pushing by feminist internet research, and opportunities for further research. The topics are access, expression, pleasure, online GBV, surveillance, data and datafication, artificial intelligence and the digital economy.

The paper also outlines how gender and related thematic areas are discussed in internet policy spaces, with a focus on the global Internet Governance Forum (IGF), and offers recommendations for further research, ways of making and disseminating research, policy making and accountability, and funding priorities in accordance with the research.

“As a progressive ideology, feminism re-shapes and updates itself to respond to diverse experiences, theoretical knowledge, and interpretation of the world by less privileged communities, specifically women and queer and gender-diverse individuals and collectives.”

– Tigist Shewarega Hussen

 

What is a feminist internet and how is it different from a gendered approach to the internet and technology? The simplest explanation would be that gender and gender equality are important aspects of a feminist internet, but not the only aspects. Feminism strives for equality for all genders while recognising that none of us lives single-issue lives and our lived experiences are being shaped by one or more of the intersections we occupy such as gender, race, ethnicity, class, caste, sexuality and disability.

A feminist approach to the internet is based on a consciousness that the inequalities in our world, including the internet and other technologies, are rooted in structures and systems of power such as patriarchy, racism, colonialism and neocolonialism, militarism, authoritarianism, ethno-religious nationalism, and macroeconomic policies and business models based on neoliberal capitalism. A feminist approach to the internet champions the right of women, girls and LGBTIQA+ people to use the internet in free, pleasurable and liberating ways and the freedom, capacity and resources to design and make an internet that puts them at the centre with autonomy and agency.

Feminist engagement with the internet is as old as the internet itself and feminist engagement with technology goes back even further. Web 2.0 or the social and participatory web of the early 1990s saw the rise of platform capitalism, which over time created walled gardens and domination by a handful of corporations. In turn, there was a need for rigorous feminist consciousness in how we use, create, critique, research and engage with the internet. Over the years, research has shown that women’s and girls’ increased access to the internet is directly proportional to the increase of violence against them online.

During the COVID-19 pandemic, we are seeing this more clearly than ever with women, girls and LGBTIQA+ people using the internet with greater frequency due to the mobility challenges of the pandemic, while experiencing an increase in online violence. A feminist approach to the internet shows that these correlations are not only due to patriarchy and gender inequality but also a number of other contributing factors, including but not limited to:

• The capitalist business models of internet platforms that consider any engagement to be profitable engagement.
• The colonial and colonising design and infrastructure of the internet.
• A lack of transparency and accountability from “those who develop and implement powerful technologies.”
• A lack of access to justice for marginalised groups of people.

There are various formations of feminists from different social movements imagining and making a feminist internet, from grassroots and community- based groups like EnRedadas Nicaragua and Metras Palestine, to national and regional groups like Derechos Digitales and the Digital Rights Foundation, to transnational and global formations like the FIRN and the loose collective of feminists holding space together as the feminist internet network, including those who have come together to create and build on iterations of the Feminist Principles of the Internet. These imaginations and makings of a feminist internet come from Africa, Asia, the Middle East and North Africa, Latin America and Eastern Europe, and are challenging the hegemonic power that North America and Western Europe hold over sources of knowledge and the skin and bones of the internet. Making a feminist internet challenges the neoliberal capitalist patriarchal articulation of the internet and looks to support and resource more girls, women, people with disabilities, LGBTIQA+ people, low-income groups of people, etc. to make and inhibit our own feminist infrastructure.

Feminist principles of the internet are a great entry point to exploring a feminist internet. They are “an expression of the kind of internet we would like to have, and will contribute to shaping” which works towards “empowering more women and queer persons – in all our diversities – to fully enjoy our rights, engage in pleasure and play, and dismantle patriarchy.” The crosscutting thematic areas of research explored in this white paper are a documentation, critical analysis and contribution to the building blocks of a feminist internet.

“Feminist research is messy. I’m grappling with just how messy it can be; but also how human it is...”

Research is an important component of imagining and making a feminist internet. As noted by the FIRN, “Building evidence, collecting data, visualising and effectively communicating research findings are critical and important ways to influence and impact activism as well as policy reforms,” and this is what feminist internet research sets out to do.

Feminist internet research applies a critical feminist lens to every aspect of research, from methodologies to substantive analysis to disseminating knowledge to citing and building on each other’s knowledge. It “focuses on knowledge building as critical feminist intervention by paying attention to multiple and intersecting structures of power that exist online and on the ground.” Feminist internet research aims to identify communities, experiences and needs that are usually marginalised in internet policy discussions and decision making, and ensure they are centred in how the internet is shaped, experienced and governed. Feminist internet research also grapples with the role, power and privileges researchers hold in making knowledge about and in collaboration with such communities. The following statements are some reflections of feminist internet researchers on their research as feminist praxis.

 

Feminist research is messy. I’m grappling with just how messy it can be; but also how human it is...

If you don’t pay attention to the power dynamics that feminism exposes and puts in front of the research you’ll miss something important. My learning is that it’s not enough to talk about human rights, and that feminism can provide me with a proper lens regarding research and the internet.

I’m learning that there is a big – and hugely productive – silence between the process of the research, what you are learning, and what is being written out in the text that someone else engages with. Reading that distance and silence is an act of active reading.

FEMINIST RESEARCH METHODOLOGIES

The repository that accompanies this white paper names and elaborates on feminist research methodologies used by researchers. While it is too dense for the purposes of this paper, the outline below presents the methodologies designed, used and reflected upon by a sample of eight research projects from Africa, Asia, the Balkans and Latin America. These represent the current body of research produced in partnership with the FIRN.

While all eight research projects used qualitative methods as their primary research methods, five used mixed methods that also included quantitative methods such as surveys and questionnaires. However, some of the surveys were carried out through in-person meetings with participants so that researchers could build trust and also provide support where needed. Some of the qualitative methods used included in-depth interviews, participatory and immersive workshops, and focus group discussions.

 

Most of the researchers had intentionally designed the research projects to be grounded in feminist ethics of care, constantly questioning the power imbalances between the researchers and research participants and trying to address them in various ways.

India

The Centre for Internet & Society (CIS) considers its class and caste positionality in researching platform workers who are mostly Dalit, Bahujan and Adivasi women from low-income groups. One way they try to acknowledge and address this is through their close collaboration with the Domestic Workers Rights Union (DWRU) as the co-researcher in the project. They reflect that this was also critical in ensuring that the research was designed and implemented in ways that were accountable to the workers who participated in the research.

Bulgaria

The BlueLink Foundation discusses measures they took to try to ensure that research participants who were survivors of online GBV did not face any further violations while taking part in the research – not just privacy and safety violations, but also emotional violations. They also discuss the toll that some of the discussions took on them as researchers, given that they also belonged to the same communities as some of the participants.

Kenya, Ethiopia, Uganda, Senegal and South Africa

 

Pollicy shares that all research implementers went through training on ethics before data collection began and that, among other things, it was emphasised to participants that there was no obligation to share any information that they were not comfortable sharing about their experiences. Whenever a participant needed psychosocial support after an interview, the researchers were able to connect them to support services.

Brazil

Vedetas, which carried out an action research project on community networks, has many reflections on how they immersed themselves in the community they researched, including home visits in the community when they arrived, open and sometimes uncomfortable discussions about race and gender dynamics between the researchers and participants, scheduling discussions while paying attention to school schedules and childcare, offering collective meals to participants to ensure fewer disruptions of cooking and housework, and others.

This section shares findings from a thematic review of research that attempts to embody some of the visions, politics and praxis of a feminist approach to the internet that were discussed earlier.

The research collected in the repository covered a wide range of topics. Each of them cut across one or more of eight overarching thematic areas which are reviewed in this section.

Access

WHAT IS A FEMINIST APPROACH TO ACCESS?

 

What is meant by “access”? What would “meaningful” access entail and how can research, policy advocacy and decision making on access and gender go beyond “inclusion” as an indicator of success? These were some of the questions explored in an earlier mapping of feminist internet research which found that there is “an increased awareness that mere access is insufficient” in the research surveyed. A feminist approach to access aims at “enabling more women and queer persons to enjoy universal, acceptable, affordable, unconditional, open, meaningful and equal access to the internet.”

INFRASTRUCTURE AND DESIGN

Feminist internet research on infrastructure shows there is much scope for improvement and even overhauling inbuilt issues in the conceptualisation, design, building, deployment and maintenance of infrastructure – whether hardware, software or digital networks – that can limit how women and other marginalised groups of people access and use the internet.

• There is a need to understand social media use in terms of a “platform ecosystem” rather than single platforms, which “can reveal greater insights for designing technology to support privacy and participation on social media” that would enable greater and safer active use of the internet, an important aspect of access.
• Infrastructure can limit the ways groups of people such as people with disabilities access and use the internet.
• Community-based autonomous infrastructure can be “access as resistance” although with gendered and intergenerational power relations in setting up and maintaining community networks

ADDRESSING THE DIGITAL GENDER GAP

This white paper finds continued focus on the “digital gender gap”, underlying causes and strategies for reduction. These are anchored to various frameworks including development, rights-based, feminist and justice frameworks.

• The gender gap in mobile phone ownership remains an issue in Africa, Asia and Latin America.
• Cultural and social norms around gender need to be considered when envisioning access for women.
• There is a need to consider legacies of colonialism that many countries continue to grapple with and explore access within “an anthropophagist model for the re-appropriation of the colonial assets by the oppressed.” There is “a strong link between low female labour force participation, access to disposable income, perceptions of women’s need for the internet and women’s actual access to and use of the internet.” There are also various challenges related to affordability as a barrier to access.
• “Women’s digital citizenship” can be a comprehensive framework that could counter the reduction of gender and digital rights to merely addressing the access gap and instead conceptualise access as an end in itself rather than an ongoing process.

FROM ACCESS TO ACTIVE USE

Access is an ongoing process and we must consider not only the factors that affect access and use but also the factors that affect the retention of such access and use.

• Cost and speed of internet, lack of time, security, privacy, lack of local language content, etc. can affect active internet use. As can device ownership, affordability, and a lack of awareness and skills about the internet cited more often by women.
• It is critical to understand women’s internet access and use beyond “passive consumption” and to fully consider women’s role, desire and potential to be “ICT producers and designers”.
• Barriers to internet use and access can be increasingly attributed to the state, affecting the economic, social, political and cultural aspects of women’s lives in particular.
• While resistance to internet shutdowns points out economic ramifications of shutdowns, these often do not take into account the effects on the informal economy that many women and other marginalised groups of people are part of.

GENDER IDENTITY IN DISAGGREGATED DATA ON ACCESS

Often the gendered dimensions of internet access and use are problema- tised by considering the experiences of women. This is partly due to the reliance on quantitative data such as household surveys or broadband statis- tics when it comes to researching access which are often not disaggregated, and even when disaggregated by gender, it is often by sex or a gender binary. While some research acknowledges this as a limitation, it is one that feminist internet research needs to grapple with.

EFFECTS OF COVID-19 ON ACCESS

From 2020, a number of researchers have considered the gendered effects of COVID-19 on internet access and use.

• While there has been an increased need to use the internet during the pandemic, “many women, who previously accessed the internet outside of their homes, like at educational institutions and in workplaces, libraries, etc. are unable to do so due to the lockdown and mobility restrictions that have been put in place due to the coronavirus.”
• While education has shifted mostly online during the pandemic, most women and girls are facing challenges around connectivity, devices and freedom to be online.
• Disease surveillance for COVID-19 is becoming all-encompassing and undermining people’s privacy, bodily integrity and dignity.

EXPRESSION

WHAT IS A FEMINIST APPROACH TO EXPRESSION?

According to van der Spuy and Aavriti, “The potential of ICTs for opening up new public spheres or platforms for all kinds of expression – including political, religious and sexual expression – is a relatively moot point.” Several dimensions of expression have emerged from feminist research, such as the criminalisation of expression and the various ways in which women, LGBTIQA+ people, people living under conservative regimes, etc. use the internet to subvert norms. A feminist approach to expression online tries to understand ways in which expression, including gender and sexual expression, is limited by various forms of violence (both online and on-ground); notes the importance of expression towards finding community and building connections; and seeks to provide critical analysis of laws and policy making on sex, sexuality, sexual orientation and gender expression. It “prioritises the amplification and unrestricted expression of our narratives and lived realities, as well as resisting and fighting against state and non-state actors’ various efforts against our freedoms, agency, and desires.”

EXPERIENCES OF EXPRESSION

• “Expression around sexuality accelerates reactions,” and while these reactions can include pleasure, community, activism, etc., they can also present backlash, violence, censorship, self-censorship, etc.
• These reactions are often mediated and/or capitalised on by the same platform economy in which such reactions take place.
• The cis and trans women and non-binary people who participated in the research all shared that to exist on the internet as themselves is to risk being harmed, and that expressing themselves online in their public and political lives is always shaped by and viewed through societal expectations and cultural values. A study from Brazil makes similar observations, noting that “it is within the cultural and social sphere that ideas around the inferiority of women are upheld and reinforced.”
• There is a “tension between law and morality” and a possible feminist faultline in using the law and the state to regulate the internet and to find redress for online violations. Approaching the courts about freedom of expression online may end up making women “subjects of state control”.
• There is the potential in online expression for gender and sexuality to escape fixed identities that might be assigned to us and instead be “ambiguously visible”.
• The nexus between violence and expression needs to be considered more in feminist internet research

CRIMINALISATION OF EXPRESSION

• Often separate laws are enacted and applied to regulate freedom of expression online which, as with offline laws, interpret all sexual expression as “obscene, immoral or indecent”.
• Laws and regulations that are intended to protect freedom of expression often end up being used to censor, silence and even criminalise. The intersection of race is of particular importance.
• Feminist internet research continues to make an important link between GBV and the criminalisation of expression, particularly sexual expression. Even when the law takes a progressive position on sexual expression, it is done from the angle of free speech and expression rather than consent or dignity.
• There is a growing need for feminists to research and understand how expression manifests, and is criminalised, within new online cultures, especially in relation to social class.

PLEASURE

WHAT IS A FEMINIST APPROACH TO PLEASURE?

This white paper understands pleasure as a crosscutting theme but creates a standalone space for it in the analysis. A feminist approach to pleasure is fundamentally different from neoliberal capitalist conceptualisations of pleasure as a commodity and an aspiration. A feminist approach recognises pleasure as central to concepts of justice, whether economic, sexual and reproductive, disability, data, design, etc., and something we are inherently entitled to experience. For example, putting pleasure at the centre would reframe how we approach digital security; “security makes you small, pleasure is expansive – how do you think about security from the point of view of pleasure?” In other words, approaching various aspects of the internet such as access, expression, infrastructure, security, etc. through a framework of pleasure would make our thinking and policies more expansive and imaginative rather than reducing people to mere economic subjects or to just one identity out of the multitudes they hold.

PLEASURE ON AND THROUGH THE INTERNET

• The internet has potential to reconfigure desire, longing and identities, and “construct lives in spaces away from the policing offline world.”
• There is discomfort among women and queer people, including those with disabilities, in discussing or being open about experiencing pleasure, especially in relation to sexuality, due to “social disapproval”, “losing credibility as activists” and “risk of judgment”.
• Pleasure in public space can be an act of protest.
• Feminist visions and practices of pleasure, play and fun can be “online sexual disobedience” and “a feminist strategy of technoresistance”.

POSITIONING PLEASURE WITHIN INTERNET POLICY DISCUSSIONS

• Pleasure is for the most part “a rarely invoked term or idea in policy and decision-making spaces,” though often downplayed due to stigma, conservatism and discomforts with the commodification of pleasure that make it seem indulgent.
• There is considerably less discussion on pleasure and consent than on violence and harm in internet policy spaces and a need “to create more positive content around sexuality and self-expression without dismissing the focus on violence and harm.

ONLINE GBV

WHAT IS A FEMINIST APPROACH TO ONLINE GBV?

The Feminist Principles of the Internet state that online violence “such as cyberstalking, harassment and misogynist speech […] encompasses acts of GBV that are committed, abetted or aggravated, in part or fully, by the use of information and communication technologies (ICTs), such as phones, the internet, social media platforms, and email.” A feminist approach recognises such violence to be on the same continuum as violence that happens offline and on-ground and places upon online GBV the same kind of importance, urgency and need for accountability and justice. A feminist approach to online GBV calls on the state and internet platforms to address the issue, but also recognises that the neoliberal state and capitalist platforms are part of the problem. A feminist approach to violence looks to reclaim the rights and freedoms of women and LGBTIQA+ people, and is centred on pleasure and solidarity.

LINKS BETWEEN ONLINE GBV, ACCESS, EXPRESSION AND PLEASURE

• “The continuum of violence has blurred the gap between online and offline spaces, whereby violence that begins online can be continued offline and vice-versa.”
• “Articulating one’s own sexuality or taking a position around sexuality is entering a perilous sea where attack and revenge can be enacted at any moment.”
• “Online forms of violence against women and girls stem from the reluctance to accept their agency.”
• Online GBV against women and girls is continuing during COVID-19.
• The online-offline continuum of violence experienced by women and LGBTIQA+ people affects expression and access: “I stopped engaging online, just as I stopped walking down dark alleys.

STATE AND PLATFORM RESPONSES TO ONLINE GBV

Feminist research documents and analyses responses and non-responses by the state and platform companies towards addressing online GBV, and in turn builds an evidence base that is critical for shaping feminist advocacy on upholding rights of women and LGBTIQA+ people on the internet.

• While strengthening harassment laws and law enforcement remains a preferred method by the state, “there is a fine line between appropriate regulation and stifling of freedom of expression.”
• “The discursive space of law presents only a partial solution to feminist visions of justice,” but the law continues to carry a “very high symbolic significance” and remains an avenue of advocacy and relief for survivors of online GBV too.
• Feminist internet research also points out that the law is predominantly heteronormative, which excludes or criminalises the issues faced by queer people.
• Platform responses to online GBV are as unsatisfactory as state responses, and maybe even more discouraging given that often there are no or few legal obligations compelling platform companies to be accountable to their users.
• Internet platforms display a lack of commitment to addressing issues faced by African and other global South users including online GBV, which results in very low rates of reporting by users about instances of online GBV on platforms.
• “Facebook staff often lacked the necessary cultural competency and literacy in the needs of caste, religious, and gender/queer minorities” due to a number of reasons, including a lack of diversity in the local teams.
• “When intermediaries are treated as gatekeepers by the law, in a bid to reduce their liability, they will over-censor.” However, as the experiences of human rights defenders, Dalit people, LGBTIQA+ people and sex workers that are documented in this white paper show, platforms tend to arbitrarily “over-censor” content by certain groups of people anyway.
• The challenges faced by commercial content moderators are feminist issues too, including the insecurity of the work, secondary trauma from consuming offensive and violent content, lack of psychosocial support, poor working conditions, and lack of benefits.

FEMINIST RESPONSES TO ONLINE GBV

• More feminists are now discussing the limitations of criminalisation and criminal justice as a response to rights violations, and highlighting, imagining and making alternative ways of understanding, accessing and experiencing justice.
• What is needed is “a more robust understanding of what help-seeking, or what justice is for the survivors beyond reporting.”
• Even the limited digital security tools that are available expect women to fend for themselves without considering the role of communities of care and solidarity, and “few interventions are aimed at preventing primary and secondary perpetrators from acting violently in the first place.”
• Often peers hesitate to support someone experiencing online GBV because they themselves may come under attack, which in turn can decrease the possibility of bystander intervention in online GBV. However, online feminist organising can be a way to resist these hesitations and rise up together.
• Feminist internet research itself can be a feminist response to online GBV. Feminist research methodologies can create spaces and opportunities for survivors of online GBV to share their experiences and connect with each other.

SURVEILLANCE

WHAT IS A FEMINIST APPROACH TO SURVEILLANCE?

The mapping shows that feminist research from the last five years is addressing concerns around surveillance in various ways. A feminist approach recognises and critiques surveillance as “the historical tool of patriarchy, used to control and restrict women’s bodies, speech and activism.” A feminist approach to surveillance seeks to understand how privacy, surveillance, violence and related concerns affect people in diverse circumstances, the differences and links between social and state surveillance, and ways in which surveillance by different actors is subverted by the subjects of such surveillance. In the past year we have also seen feminist research and analysis emerge on the implications of disease surveillance as a response to COVID-19.

SUBVERTING AND RESISTING SURVEILLANCE

• “The effects of surveillance are never merely personal; they are structural,” and therefore a feminist response to surveillance must go beyond just privacy protections and call for “transparency and accountability on the part of those who develop and implement powerful technologies.”
• Some of the ways in which women and queer people subvert surveillance online are through anonymity, creating more than one account on platforms, and challenging notions of “digital izzat” (respect, honour) through selfies, etc.
• Resisting and subverting surveillance has also become very much a part of feminist and women’s rights organising and activism.
• “The resistance against surveillance practices needs to be delved into in intersectional ways,” bringing up questions of who gets to resist surveillance, who has no choice but to tolerate it, etc.

COVID-19 AND SURVEILLANCE

Surveillance has become a key strategy in how governments around the world are addressing COVID-19. Therefore, feminist critiques of state and corporate surveillance have to traverse the tricky path of contending with disease surveillance with its connotations of public good and public safety.

• “While data can provide insights on safety during a pandemic, it cannot itself keep people safe.”
• Care and mutual trust are missing when surveillance becomes the core of a pandemic response and “surveillance of the disease should not be conflated with surveillance of bodies.”
• Increased and legitimised surveillance during COVID-19 has worsened the protectionist and controlling approaches towards women with disabilities.
• “Mass data collection could be accepted as the norm even after the coronavirus has become less of a threat to the public” and possibly strengthen the nexus between bio power and patriarchal power.

DYNAMICS BETWEEN SURVEILLANCE AND CARE, PRIVACY, CENSORSHIP, SECURITY, ETC.

• A number of researchers examine the dynamics and faultlines in how the state, corporations, civil society and other institutions and individuals justify surveillance, and ask what is needed to challenge them.
• There can be a “risk of function creep in use of biometrics” and marginalised groups of people can find themselves in difficult positions when their access to health and other essential services is tied to potential surveillance.
• “Online technologies that are actively policing and criminalising sex work contribute to a broader system of state-corporate funded surveillance,” which connects the dots between surveillance, content moderation and platform capitalism.
• The challenge for feminists is “adopting a lens that is cognizant of a history that proves intersectional commitments in their overall resistance against surveillance.”

DATAFICATION

WHAT IS A FEMINIST APPROACH TO DATA AND DATAFICATION?

There have been a number of feminist and intersectional engagements, and entanglements, with data and datafication in the past few years, and this section reflects on a few of the key concepts that have emerged from these. They offer ways of thinking as well as tools and frameworks as feminist internet research seeks to influence policy and governance decisions and to imagine and create our own spaces and infrastructure. A feminist approach to data and datafication examines the nature of data and constantly resists disembodiment of data. It is centred on the understanding that the consequences of data and datafication, both the harms and possible benefits, are embodied with individuals and communities facing those consequences. A feminist approach seeks to understand the effects of self-surveillance or “quantified self” developments (including the use of big data) on human rights, tries to unpack discrimination in algorithmic decision making, and introduces feminist praxis around data. It recognises “the right to privacy and to full control over personal data and information online at all levels.”

FEMINIST APPROACHES TO DATA AND CONSENT

• “Feminist perspectives on consent, in particular in the context of sexual relations” can guide changes that need to be made before and during the collection of data, a more nuanced and detailed approach to permissible uses of data, and things to be considered in relation to the data of vulnerable or marginalised groups of people.
• Accountability and transparency in data collection are required from not just the state but also development actors and non-profits in their entanglements with data.
• Queer feminist approaches to data and consent show “severe limitations of using an individualistic approach to consent as sole requirement for several interactions with our data bodies,” and view consent as a doing thing and call for it to be active, clear and intelligible, informed, freely given, specific, retractable and ongoing.

DATA JUSTICE AND DATA FEMINISM

• “Data, in its various forms, can play a vital role in feminist movements achieving the ideal transformative and just society,” but for this we must resist colonial legacies and power imbalances and “ensure that data is shared in a transparent and accountable manner and aligned with feminist principles.”
• A feminist approach to data “would require that this crisis of subjectivity and sociality be re-imagined through a radical practice of community and connectedness”­ that moves beyond capitalist and market-based conceptions of data.
• “The starting point for data feminism is something that goes mostly unacknowledged in data science: power is not distributed equally in the world,” and it seeks to “examine power, challenge power, elevate emotion and embodiment, rethink binaries and hierarchies, embrace pluralism, consider context, and make labor visible.”
• Data colonialism, datafication and resultant power and gender dynamics are “significant for Africa primarily because of its fragile economies, poor infrastructure and inadequate justice institutions.”
• There are links between internet access/infrastructure and unconnected people in Africa being considered “potential data”. Tamale notes the importance of forming and shaping “Afro-feminist alternatives” that put data justice at the centre.

FEMINIST PERSPECTIVES ON DATAFICATION

• Datafication often promotes an understanding of people and our world as the sum of data that is collected about us. Feminist researchers, scholars and activists challenge this notion by pointing out the importance of considering what is missed, hidden or cannot be sussed out of data about us.
• Datafication and related data practices can have harmful effects on embodied personhood
• “The line we draw between our body and information about our body has slowly started to disappear” and data protection has to go beyond the conventional understanding of being about disembodied information to understanding data as embodied.
• There are tensions of datafication when it occurs in research by private, academic and non-profit entities, and proposes guidelines and practices from Indigenous communities who question and challenge the ways in which their lives are datafied through research.

DATA PROTECTION AND DATA SOVEREIGNTY

• Often the data protection responses by the state are closely intertwined with sovereignty and territoriality and both could have implications on the rights of women, LGBTIQA+ people and other marginalised groups of people like migrants, refugees, stateless people, etc.
• Feminist research often frames discussions of data sovereignty and data protection in terms of the rights of data subjects.
• In order for a data protection framework to be gender responsive, its design and implementation must “consider gendered realities of the society we live in and ensure that injustices are not replicated as we race towards digital development.”
• When it comes to digitised social welfare programmes, beneficiaries, including women, should not have to “choose between privacy and social protection, food security, or a benefit that after all alleviates, but does not eliminate poverty.” Despite potential benefits, digital identification systems “have the potential to amplify harms and discrimination within society, when implemented without care and a social justice lens.”
• As a number of data protection legal frameworks emerge around the world, especially in the wake of the European Union’s General Data Protection Regulation (GDPR), feminists are responding with much- needed critique and analysis of these frameworks such as “vague language and definitions”, impact on small and medium enterprises, and the neocolonising effects of forcing other countries to adopt frameworks to comply with the GDPR that would only apply to personal data on the internet.
• Feminist critiques show the need for states to go beyond GDPR compliance and create bottom-up data protection frameworks that shift from individuals versus markets to individuals and communities versus structures and systems of oppression.
• Data sovereignty is essential for such a shift, “a potent framework” to counter and resist the kind of data colonialism that forms and strengthens “modern forms of hegemony of big tech companies.”
• A meaningful concept and practice of data sovereignty that promotes people’s rights would need to shift from treating data as a mere resource “that is simply out there,” ready to be mined, to “an extension of our bodies, even a part of it.”

ARTIFICIAL INTELLIGENCE

WHAT IS A FEMINIST APPROACH TO AI?

“Are we all equal in the eyes of AI? What are the opportunities and challenges for marginalised groups in society with AI? What can be done to ensure social justice for marginalised communities?” Chenai Chair asks these pertinent questions that are fundamental to a feminist approach to AI. Other questions might include who gets to define intelligence, what kinds of intelligence, what can the ruling class and capitalists use or abuse AI for. A feminist approach to AI challenges the singular “silver bullet” narrative about AI and big data, calls for accountability for harms caused by algorithms and AI, doesn’t deny the potential of AI for public benefit but flags that it remains a tool of neoliberalism that currently replicates the inequalities and marginal- isations of our world, and fights against algorithmic determinism while imagining and proposing ways to rebuild and transform AI.

GENDERING AI

• The popular culture representations of these “female AI” put a focus on sexuality and are expected to and indeed do sexual and emotional labour for the male protagonists.
• The gender imbalance in the development of AI can be a possible reason behind such “gendered tech narratives”. Globally, less than 25% of AI professionals are women.
• The roots of AI as a concept have been masculine and heteronormative from its inception, with a group of men getting to define intelligence in the context of AI, who and what purposes it will serve, and what problems it will solve. So attempting to expand AI to be feminist without reimagining AI may prove futile in the context of systemic inequalities feminists struggle against.

ALGORITHMIC FAIRNESS

Feminist research is documenting, analysing and theorising on how decisions made using unfair algorithms and systems result in further marginalising and excluding people who are already marginalised, including on the basis of gender, race, income, class, etc. Complementary to this, and equally important, is research that addresses existing and ongoing challenges in algorithms and classification systems that result in such decisions, marginalisation and exclusions.

• Big data systems intended for the global South are framed, designed and deployed as instruments of empowerment within an “inclusive capitalism” framework. “Alternative modes of inclusivity should be sought beyond the default neoliberal approach.”
• Machine-learning algorithms perform with less accuracy when it comes to gender and race and there is a need to evaluate bias in such algorithms and accompanying datasets. However, improving such inaccuracies, while useful in the short term, may not address the power structures that AI is based on and is reinforcing.

AI GOVERNANCE

As AI and algorithmic decision making permeate everyday life through governments, corporations, educational institutions, medical institutions, etc., the calls to govern AI have intensified.

• Dominant narratives on AI governance focus on aspects such as privacy and other human rights violations, the future of work, opaque decision making, and reinforcement of societal biases based on gender, race, income, etc. that result in discrimination.
• A feminist approach to AI governance raises that these dominant narratives often do not “fully address the entanglement of AI in neoliberal capitalism and what this means for the life-chances of individuals and communities.
• Feminist engagement with AI raises that accountability in AI governance should not be limited to fixing AI but about all states having the right to scrutinise AI and algorithms that affect them and demand transparency from both state and non-state entities that create and use AI.
• The challenges and opportunities posed by AI can vary, including between the global North and the global South, and binding normative frameworks for AI accountability need to make space for such complexities.
• AI governance is also important in the context of development. Researchers note that “a promise for many is that big data and AI will augment statistical measurement” but that the ability of such data “to represent fairly everyone in the society is questionable and any solution created from this data has the potential to be inequitably deployed.”

FRAMEWORKS TO DECOLONISE AI

• A feminist approach to AI looks to dismantle the power structures that AI is based on and is reinforcing.
• “Beyond even a human rights framework, decolonial and transfeminist approaches to technologies are great tools to envision alternative futures and overturn the prevailing logic in which AI systems are being deployed.”
• AI projects and tools can be assessed through a decolonial feminist framework that is embedded in values such as “agency, accountability, autonomy, social justice, non-binary identities, cooperation, decentralisation, consent, diversity, decoloniality, empathy, security, among others.”

DIGITAL ECONOMY

WHAT IS A FEMINIST APPROACH TO THE DIGITAL ECONOMY?

A feminist approach to the digital economy commits to “interrogating the capitalist logic that drives technology towards further privatisation, profit and corporate control” while also constructively creating “alternative forms of economic power that are grounded in principles of cooperation, solidarity, commons, environmental sustainability, and openness.”

“What might be rich areas for our anger, critique, celebration and future research on digital economics and how might we bring perspectives from the global South to these conversations?” Some of the areas highlighted in response include unpaid digital labour, digital skills, entrepreneurship, the platform economy and the environmental impact of technology use. It is heartening to note that many of these areas and more are being researched, documented and reimagined by feminist internet researchers.

GENDER EQUALITY IN THE DIGITAL ECONOMY

Gender inequalities and disparities in the digital economy is a widely researched topic and this section discusses some of the key observations emerging from it.

• “Economic relations and economic activity are being reorganised through the advent of platforms and databased intelligence” and this is resulting in adverse effects and challenges for women and other marginalised groups of people while also creating positive effects and opportunities for them.
• Addressing the challenges and seizing the opportunities require structural reform and “institutional transformation from global to local levels.”
• The infrastructure underpinning the digital economy, from microchips to server farms, are based on “an unsustainable exploitation of natural and common property resources in the global South”­ that threatens access to water, food security, etc. and in turn threatens lives and livelihoods of the most marginalised.
• Informal and unpaid work is gendered and invisibilised and digitisation of informal work and unpaid labour can then render such work invisible in the digital economy too.
• Reconceptualising a feminist world of work in the digital economy includes addressing gender stereotypes and underlying social and cultural norms, more gender-disaggregated data, and “new thinking for revamping legal institutional mechanisms at national and subnational levels.”
• The digital economy is possibly a feminist frontier to reimagine the concept of work itself. “We are not necessarily talking about ‘machines replacing humans’ but rather displacing traditional work roles and thus calling for a re-imagination of human work.”

PLATFORMISATION OF WORK

Work is increasingly being mediated and organised via digital platforms, and feminist researchers are exploring the implications of this. Especially given that most such platforms operate on capitalist and extractivist business models although there are other models such as cooperative platforms also gaining ground.

• Western scholars argue that the promised “flexibility” of platformised work hides various costs and rules involved in such work but this argument is not as straightforward when considering platformised work in the global South, “where informality is rampant” and technology provides opportunities to be part of the market.
• Even though there is increasing demand for Africa as a market for outsourcing digital labour on platforms such as Fiverr, Upwork and Amazon Mechanical Turk, this is not just due to increased internet connectivity in the region but also the rising labour costs in dominant outsourcing markets like India and the Philippines
• Backlash against platform workers unionising to demand better labour protections and working conditions shows the need for fresh approaches that include organising across countries and regions.
• Ride-hailing or ride-sharing applications, both global and local, can end up reinforcing “the existing unjust institutional and social infrastructure.” “Platforms rely on and amplify unequal structures of power that workers already experience.”
• Gender, class and caste influence how drivers and customers interact with each other while mediating via platforms, including the sexual and physical safety of women customers.
• Platformisation of domestic work is another critical area of research. Research finds “gig economy induced challenges, occupational challenges, and occupational challenges exacerbated by the gig economy nature.”

UNPAID CARE WORK AND THE INTERNET

• Internet access and digitisation are not overnight transformations for women and other marginalised groups of people. While platform-based work “provided women flexibility, shorter work periods and better pay, it has not actually freed them from their household responsibilities.”
• While paid and unpaid care work is performed mostly by women, “there have been no policy discussions to date regarding the relationship between paid and unpaid care work.”
• The flexibilities afforded by digital technology may not be transformative if the challenges in managing unpaid care work including inadequate care facilities for children and elders are not addressed in parallel.

SEX WORK AND THE INTERNET

Sex work (on the internet, mediated through the internet and outside the internet) is a microcosm for a number of topics discussed in this white paper such as access, expression, pleasure, violence, surveillance and datafication. Locating sex work and the internet under the larger topic of the digital economy was a deliberate decision because understanding, recognising, decriminalising and supporting sex work as part of the digital economy is critical to addressing a number of challenges sex workers face on and off the internet.

• Websites used by sex workers were being shut down in a way which is compared to “tactics akin to those used for ‘cleaning up’ sex workers from public view” and such restrictions to access to technology are not just censorship but also “a form of structural violence”.
• When corporate entities like platforms arbitrarily set community standards that restrict or ban sexual expression including pornography made independently, consensually and ethically, it can result in self- censorship by performers to present a corporate, sanitised, edited version of their sexuality that is compatible with the market-driven architectural infrastructure of privatised platforms.
• Sex workers noted that if your livelihood is dependent on the internet, then internet shutdowns end up reducing or completely restricting your monthly income.
• COVID-19 is also having an impact on how sex workers use the internet for their livelihood, with staying confined to home creating challenges in communicating with clients, especially when families are not aware of a sex worker’s profession.
• While digitisation has indeed helped sex workers with devices and internet access to find safer and more ways to do sex work, being visible participants in the economy can come with risks, especially when soliciting and living off the income of sex work are criminalised. This shows the intersecting complexities of access, data, surveillance and the digital economy for sex workers and why their voices, concerns and needs need to be at the centre of policy decisions.
• Sex workers also need support with their digital safety and security in order for them to be active in the digital economy while navigating these complexities. For example, “setting up web pages and creating contents to free servers, always focusing on the use of free and secure tools.”
• “Taking a sex worker lens to tech is not simply about involving more sex workers in tech design” but also raises “fundamental questions about ethical design, participation, access, privacy, surveillance, violence, and re-visioning new worlds.”

A FEMINIST FUTURE OF WORK

Feminist internet researchers are drawing out complexities and nuances of the digital economy and questioning neoliberal capitalist macro- economic policies while also laying out pathways for the future. This section briefly outlines some of those ways forward.

• A feminist approach to the digital economy commits to “interrogating the capitalist logic that drives technology towards further privatisation, profit and corporate control” while also constructively creating “alternative forms of economic power that are grounded in principles of cooperation, solidarity, commons, environmental sustainability, and openness.”
• The environmental impact of technology is an important topic of discussion as we research and expand the contours of feminist digital economics.
• “The beneficial effects of ICTs – reducing energy consumption and facilitating the shift towards renewable energy – need to be weighed against the direct detrimental effects of our change to a digital economy.
• A degrowth approach might be the way forward, and “with all its infinite challenges, degrowth can be a concrete stimulus for technologists, civil so- ciety, academia, governments and companies to move away from an ex- tractivist logic and shape a sustainable digital economy.”
• A feminist future of work is not about reskilling people or bringing them into already exploitative structures but about exploring and championing alternative forms of economic organisation such as cooperatives.

“Feminist and human rights engagement with these spaces shows that our issues continue to be marginalised in such spaces.”

This section is a brief outline of if and how the thematic areas discussed so far – and an intersectional feminist analysis of them – have emerged in the internet policy space, with a focus on the global Internet Governance Forum (IGF). This outline is by no means exhaustive and is also cognisant of the relevance of the IGF continuing to diminish in how policies to govern the internet are actually made by various stakeholders. However, it acknowledges the advocacy done by feminist and human rights activists in the IGF space over the years, takes stock of if and how specific issues have been discussed at IGFs, and attempts to shed light on how feminist internet research needs to be positioned in internet policy and governance discussions.

Who governs the internet? Ideally, a wide range of actors who can address the varied and often crosscutting aspects of the internet such as the technical, human rights, legal and academic aspects, while centring the digital sovereignty of people. However, as the thematic discussions in this white paper have illustrated, the internet seems to be governed in various and often haphazard ways, with platform companies and other corporations driving a neoliberal logic of the internet, and governments applying existing laws as well as introducing internet-specific laws that are more often used to crack down on dissent and consolidate state power rather than protect internet users.

There is a range of geopolitical and multistakeholder gatherings, organisations and mechanisms – albeit with increasingly less space for civil society stakeholders and with states and the private sector consolidating power and resources in them – that attempt to discuss, negotiate and implement internet governance. But feminist and human rights engagement with these spaces shows that our issues continue to be marginalised in such spaces. There is also a noticeable gap between internet governance spaces and the policy making by governments and companies (such as content moderation guidelines of platform companies), which shows that the state and corporate hegemony over the internet and its governance needs to be contended with in order for many of the recommendations made by feminist and rights-based internet research to be implemented in meaningful ways.

Feminist engagement with internet governance and policy making goes back a long way and continues to happen at national, regional and global levels, in various forums and spaces. Gurumurthy maps these engagements such as the Fourth UN Conference on Women in Beijing in 1995, the Millennium Declaration of 2000, the formation of the World Summit on the Information Society (WSIS) in the early 2000s, various ICT for development (ICTD) initiatives over the years, the IGF, and the 2030 Agenda for Sustainable Development. As with the development sector at large, these histories also show how feminist analysis of the internet got subsumed into the development framing of women’s empowerment and the ongoing work by feminists to address this through a comprehensive approach to internet governance that considers gender justice and economic justice in tandem. Gurumurthy observes that feminist activism in internet governance spaces has resulted in significant progress, such as raising issues around surveillance and privacy, online GBV corporate accountability, internet intermediary liability, etc., but notes remaining gaps and challenges.

Jensen points out that “from a feminist perspective, the institutions (of internet governance) vary tremendously in terms of the possibilities for meaningful involvement, from relatively open setups such as the IGF to relatively closed ones such as the ITU [International Telecommunication Union].” The IGF is a platform for stakeholders interested and invested in internet governance and policies to come together and share information, knowledge and best practices and have critical discussions on internet policies and policy recommendations. While the discussions and outcomes of IGF proceedings are not binding on states or other participants, it is one of the few forums that engages a broad cross-section of actors from government, the private sector, civil society and academia at national, regional and global levels (subject to feminist critiques such as those of Gurumurthy, who notes that “multistakeholderism in the IGF sense makes challenging neoliberalism and its mutations – essentially an antagonistic politics – in countries dependent on global IT capital highly difficult”). For the purposes of this white paper, the focus will be on the IGF at the global level, given its relative openness as Jensen mentions, the broad range of actors, as well as robust documentation of how feminist and gendered engagements have played out in the space over the years.

In 2020, a stocktaking exercise undertaken by the IGF Best Practice Forum on Gender and Access tried to understand progress and gaps in how gender is discussed at the IGF.

A summary of recommendations that emerged from the Best Practice Forum stocktaking exercise follows:

• Topic(s) of gender should be mainstreamed at the IGF. While there should still be sessions focusing on gender, other sessions also need to be intersectional (on gender and beyond) in the substance of the discussions as well as the diversity of speakers and participants.
• The discussions on gender at the IGF should expand beyond topics of violence and harm and also integrate expression, pleasure, consent, etc. in order for a more comprehensive and dynamic thinking and recommendations to emerge from this policy space. There is a need for more gender diversity at the IGF along with more disaggregated data to measure progress towards this. The participation of women and gender-diverse people at the IGF should not be tokenised and there should be space, resources and sustainable plans for them to meaningfully engage with internet governance processes.

It is encouraging to learn from this stocktaking exercise that there is increasing interest and commitment to integrating gender into discussions at the IGF. However, the feminist internet research mapped for this white paper from the same time period shows that there is scope for IGFs to become more ambitious and complex in their discussion of gender. Feminist internet research on data, datafication, AI and the digital economy is particularly rich, and as this white paper has demonstrated, all these themes are interlaced with, and add nuance to, discussions around gender and access. The recommendation to go beyond discussions of violence and harm is commendable too, and is in line with some of the con- clusions and recommendations emerging from this white paper.

Another observation is that while some of the challenges and policy shortcomings noted in the feminist internet research mapped here have been raised in the internet governance space, there does not seem to be an effective feedback loop from the IGF to policy makers at national and local levels as well as no accountability mechanisms or political will to implement the findings and recommendations emerging from both research and multistakeholder IGF discussions. The onus of fixing this gap should be on policy makers, but civil society and academia have an opportunity here to ensure that findings and recommendations from feminist internet research are also submitted beyond internet governance spaces to other forums, like treaty body reviews (for example, the Convention on the Elimination of All Forms of Discrimination against Women, the International Covenant on Economic, Social and Cultural Rights, the Convention on the Rights of Persons with Disabilities, etc.), UN commissions, the International Conference on Population and Development, the Beijing Declaration and Platform for Action, the 2030 Agenda on Sustainable Development, advocacy around a binding treaty on business and human rights, etc.

Based on the conclusions reached by this white paper as well as the thematic review of feminist internet research on access, expression, pleasure, online GBV, surveillance, data(fication), AI and the digital economy, the following are the recommendations for how to achieve further progress in and through feminist internet research.

This can also serve as a starting point for those who want to adopt a feminist approach to the internet and development.

FURTHER RESEARCH

• More feminist research on and by marginalised groups of people from the global South that unpacks their experiences, needs and imaginations of the internet, including how people are building and sustaining resistance and alternatives.
• Research, especially feminist action research, that approaches access, infrastructure and design as locations of feminist resistance, including feminist autonomous infrastructure.
• Critical analysis of neoliberal capitalist conceptualisations of access such as “Free Basics” and how they affect digital citizenship.
• Feminist conceptualisations of access and universal access that imagine and create possibilities of a free and autonomous internet.
• An intersectional analysis of expression that goes beyond gender and sexuality, and engages with how class, caste, race, religion, indigeneity, etc. influence and affect expression.
• Understanding our experiences of the internet as embodied and human experiences that have embodied and human consequences.
• The discussion on pleasure, understanding it as a vital part of the continuum between access, expression and violence. There are various entry points to discussions about pleasure including but not limited to sexual expression.
• How economic injustice affects and drives online GBV, surveillance, datafication and the control and restriction of sexual expression.
• Links between the digital economy and online GBV.
• Feminist conceptions of justice in relation to the internet, and the faultline feminists have to occupy while demanding state intervention in addressing violence, both online and offline.
• Comprehensive, community-based and collective interventions beyond criminal justice for rights violations that happen on and via the internet, which take into consideration the economic, environmental, emotional and other aspects of harms and justice.
• Narratives around the quantified self, digital labour aspirations, local platform economies, etc. and their complexities and depth in the global South.
• Research on gender and the digital economy, platformisation in particular, that considers the various ways in which power is consolidated and distributed, including class and caste.
• Consideration of the gendered impact of internet shutdowns that critically analyses the state and development discourse on internet access and women’s empowerment.
• Collective, community-based and non-linear conceptualisations and practices of consent.
• Ways of making and disseminating feminist internet research.
• Researchers, activists and other civil society actors need to track how feminist internet issues are raised and discussed in internet policy and governance spaces, and if and how they are integrated into policies.
• Research data policies must allow for anonymised datasets to be made publicly available for those who want to further study or analyse them, including through feminist and decolonising frameworks of analysis.
• Sufficient resources must be made available for researchers to design and fully implement feminist research methodologies that centre the care and well-being of research participants, as well ensuring that they are able to conduct follow-up and provide accountability to the communities who were part of the research.
• Feminist internet research must be translated into and/or be produced in local languages for country contexts in which communication with policy makers and the public is more effective in local languages. This would also open up space for more local and grassroots groups to be involved and encouraged to participate in knowledge creation and sharing on feminist approaches to the internet.
• Feminist internet research must be published in open access formats that are accessible for various levels of connectivity, devices and abilities, and be accompanied by clear guidance for citation (such as adding the preferred citation to publication details) in order to facilitate easier referencing.
• Internet policy spaces must have more robust, intersectional and complex discussions and outcomes on gender and intersectionality, and ensure that the integration of gender into discussions as well as stakeholder representation at these spaces is not tokenised and is instead undertaken in long-term, sustained and well-resourced ways.

FUNDING PRIORITIES

• There is a need for continued resourcing and supporting of feminist internet research formations and journeys, especially collectives that are led by and organised around the struggles of workers, LGBTIQA+ people, and minorities on the basis of caste, ethnicity, religion, indigeneity, etc.
• Cross-movement, cross-regional and transnational research and solidarity building needs to be resourced and encouraged, especially but not limited to addressing issues of extraterritoriality and pushing for multilateral accountability mechanisms.
• Long-term sustainability is needed for feminist internet research in order to enable researchers to continue their engagement with topics, participating communities and action research work; exercise feminist ethics of care; and build a stronger evidence base for the benefits of feminist internet research.