Seventy-five percent of the world’s online population is from the global South, and nearly half is projected to be women. Yet public knowledge on the internet - exemplified by Wikipedia - is primarily constructed by (white) men from western Europe and North America. One in ten Wikipedia editors are estimated to self-identify as female. In other words, the internet of the majority is produced by the minority. But Wikipedia is only one example of the deeply skewed experience of the internet: from the design and architecture of the internet, to the production and reproduction of knowledge on the internet, this globalised “public sphere” not only reflects the structural and representative inequalities of our world, it can, in many ways, amplify and deepen them. Still, the internet’s socio-technical nature can also engender potentially emancipatory processes in which communities on the “margins” of both the physical and virtual worlds can produce and curate their own knowledge online.
Whose Knowledge? is a global, multilingual campaign that aims to make public knowledge and the online experience less white, male, straight, and global North in origin. Using Miranda Fricker’s concept of epistemic injustice and Linda Tuhiwai Smith’s exhortation to decolonize our methodologies, the Whose Knowledge? campaign has supported marginalised communities like Dalits from India and the diaspora, queer activists from Bosnia-Herzegovina, and the Kumeyaay Native American Nation, to begin centering their knowledge online. The authors are feminist scholars, organizers, and technologists, and we describe the practices of decolonizing ourselves in this effort, in particular our approach to embedding feminist and anti-colonial values as we decolonize design, process, and metrics. We offer these possibilities and provocations for thinking further about a future feminist decolonized internet(s).
As a Dalit software engineer, being in a space that empowered us to create our own content was deeply inspiring. I felt I had ownership over a story so seldom told by others. I found histories and stories that were buried and forgotten. These are stories that get muted in the loud barkings of mansplaining. I learned about people like me. I learned about Grace Banu, another Dalit Software Engineer, who went through her computer science program without owning a computer. The editathon told me that I mattered, and that I had power over my own voice. I hope to continue to support more editathons in the future.
—Selvi2017, new Wikipedia editor, after participating in a Dalit History Month editing event, supported by Whose Knowledge?
Why we need feminist decolonization
The history and knowledge of marginalized communities, like the Dalit community that Selvi2017 comes from, is the history and knowledge of the majority of the world. Like many marginalized communities, Dalits (those formerly and pejoratively known as “untouchables” in South Asia’s caste system) have been denied access to educational institutions, and their histories have often been written by their oppressors instead of their own community.
In the age of the internet, unfortunately, these same biases about who writes (or can write) historical knowledge have frequently been reproduced, and sometimes deepened, online. Seventy-five percent of the world’s online population is from the Global South (“Internet World Stats” 2017), and nearly half of all women are online (International Telecommunication Union 2017). As the primary example of online public knowledge, Wikipedia is one of the most visited websites in the world (“Alexa” 2020). Yet, at last count, only 20 percent of the world edits about 80 percent of Wikipedia’s global content (Wikimedia Foundation Staff 2015), and we estimate that only one in ten of the editors is female (Galvez 2018; Hill and Shaw 2013). Public online knowledge does not represent most of the world; it does not look like most of us online.
While much of the focus of internet research and activism has been on access and usability, we think it is equally critical to examine content and knowledge production: what and whose content people “consume” or “produce” online. What do we access, what are the contexts in which it is produced, and who decides what we see?
However, to answer these questions honestly and radically, we must first ask: who are we? Without examining our positionality and reflecting on the ways in which our lives and our perceptions of ourselves and others have been shaped by structures of power and privilege, we cannot do feminist work online—in a space where so many aspects of our identities can be easily elided, according to feminist geographers Morrow, Hawkins, and Kern (2015).
This essay examines the organizational and community practices through which feminists work toward an equitable, decolonized internet. In our case, we work on centering the knowledges of marginalized communities—the majority of the world—online. But to do this in a way that moves beyond stated principles into embodied practices, we have to be able to reflect on our own intersectionalities and interpositionalities of power. We will use examples from our work to reflect upon these positions and practices. Political scientists Brooke Ackerly and Jacqui True talk about such reflection in feminist research ethics as “our commitment to self-reflexivity, our attentiveness to the power of epistemology, of boundaries and relationships into the practice of our research” (Ackerly and True 2008, 705). For us, our research, action, and advocacy need these forms of self-reflexivity.
As feminists, we know that how we do our work is as important, or even more so, than what we do. We cannot decolonize the internet without decolonizing ourselves.
Who we are
The authors of this essay are part of the leadership of the Whose Knowledge? global campaign: three cofounders and three advisors. We are academics and activists. While all of us self-identify as feminist women, we come from different backgrounds and identities: we identify variously as black, brown, white, female, from both the Global South and the Global North, cisgender, and able-bodied. We recognize that one form of shared privilege we have is the material and temporal resources that help us contribute to radical social transformation.
Whose Knowledge? is a global, multilingual campaign that seeks to make the internet—particularly public online knowledge—reflect the richness and diversity of the physical worlds we live in. We bring our feminist perspectives and politics to this work, focusing on the majority of the world, still marginalized through structures of power and privilege: women, people of color, indigenous peoples, LGBTQI+ communities, and others from the Global South. We work with marginalized community partners around the world who decide what knowledge they’d like to bring online and how. Our role is to convene, connect, and support individuals, organizations, and communities as deep allies so that the histories and stories of marginalized communities are centered online.
Why we write and how
This collective writing is part of our journey to document ourselves as we support the rich and textured online documentation of others. Throughout, we do our best to “place” ourselves—and others we reference in a substantial way—in the contexts and perspectives from which they write. As far as possible, we have used self-described identifiers of those we reference, but in other cases we have offered our own descriptions as respectfully as possible. We do this to make clear that all knowledge comes from contextual, “situated” perspectives (Haraway 1988).
As feminists, we want to address the ways in which online or networked technologies—and the “internet” itself—are deeply political. These technologies affect, and are affected by, those who design, produce, and use them. Most critically, we want to address our own self-reflexivities as scholars and practitioners in the field of feminist technologies.
Throughout this essay, we offer some of the context and inspirations in our grounded work, and some lessons learned. We divide our feminist practices into three related sections around decolonizing design, process, and metrics, and embed examples as audio files in our voices, to embody our commitment to expand nontextual modes of knowledge production.
Inspiration: Epistemic Injustice and Decolonizing Methodologies
We are inspired by the work of Miranda Fricker (an English philosopher, currently professor of philosophy at CUNY Graduate Center, in the United States, and a research professor at the University of Sheffield, in the United Kingdom) and Linda Tuhiwai Smith (a Māori scholar, currently professor of Māori and indigenous studies at the University of Waikato in Hamilton, New Zealand). We draw on Fricker’s concept of epistemic injustice and Tuhiwai Smith’s exhortation to decolonize our methodologies, bringing their intellectual contributions together here, in conversation with one another. As Whose Knowledge? centers historically marginalized communities so that they can produce and curate their knowledges online, Fricker and Tuhiwai Smith’s frameworks help us to think about a specific instance of the internet’s skew: Wikipedia. Wikipedia is not only the world’s largest collaboratively edited encyclopedia, it is also a primary example of how the best possible intentions—to create free and open online public knowledge for and from all—can still leave much to be desired in practice.
Fricker (2007, 1) uses the term epistemic injustice to describe “a wrong done to someone […] in their capacity as a knower.” She further theorizes this kind of wrong by drawing distinctions between testimonial and hermeneutical injustice. The former “occurs when prejudice causes a hearer to give a deflated level of credibility to a speaker’s word.” The latter occurs “when a gap in collective interpretive resources puts someone at an unfair disadvantage when it comes to making sense of their social experiences.” To cite two of Fricker’s examples, the police not believing a man because he is black in a white-majoritarian country is a case of testimonial injustice, while a woman who experiences sexual harassment but who lives in a culture lacking a concept for sexual harassment is a case of hermeneutical injustice. Fricker’s philosophical framework helps us to think about how different kinds of knowers and knowledge are marginalized, especially on the internet.
Relatedly, Tuhiwai Smith (1999, 1) examines how academic research is dominated by Western notions of credibility and validity. Research, she writes, “is probably one of the dirtiest words in the indigenous world’s vocabulary.” But interrogating the ways in which the West has used research to colonize peoples, spaces, and even concepts of time and knowledge can enable marginalized communities to imagine self-determination and set their own agendas. Tuhiwai Smith asks indigenous researchers and non-indigenous researchers who work with indigenous peoples to consider how they might use this awareness to decolonize their practices. Whose Knowledge? participates in this process by questioning the values embedded in the internet—largely conceived of and constructed in the Global North—and by asking how digital technologies currently both amplify and deepen structural and representative inequalities. At the same time, we believe an active lens of intersectional justice and transformation can help us reimagine and redesign the internet for potentially emancipatory work.
Principles in Practice
We recognize that one of the most crucial aspects of decolonizing the internet successfully is how well we turn principles into practice. Naming these principles is important, but we often falter in how we embody them. We believe (as Tuhiwai Smith urges us) that the dynamic process of decolonization needs a continuous (re)examination of how principle and practice are in conversation with each other.
In the context of public online knowledge, nowhere is this tension between principle and practice more obvious than on Wikipedia. We work specifically on Wikipedia because of its extraordinary reach. As one of the largest free and open information databases in the world, its content is shared across the internet in multiple ways, including through the “knowledge graph” of search engines like Google. The knowledge gaps on Wikipedia get amplified across the internet. To center marginalized knowledges in practice, one of the spaces we have to address is Wikipedia, and one of the challenges we face is the nature of its principles in practice.
Take one of the fundamental principles of Wikipedia: the “neutral point of view” (NPOV), “which means representing fairly, proportionately, and, as far as possible, without editorial bias, all of the significant views that have been published by reliable sources on a topic” (“Wikipedia: Neutral Point of View,” n.d.). This principle assumes that “neutrality” is possible and that the position and capacity of the knower are inconsequential. NPOV is similar to what Donna Haraway (1988) calls “the god trick.” Like scientific objectivity, NPOV positions facts and information as separate from and independent of contexts and knowers. Haraway herself (professor emerita in the history of consciousness and feminist studies at the University of California, Santa Cruz) describes this critique of objectivity as “feminist empiricism” (1988).
“No original research,” another of Wikipedia’s foundational principles, requires that information be documented elsewhere before it can be included in Wikipedia. Using “secondary” sources—in particular, books, peer-reviewed journal articles, and newspapers—is in keeping with encyclopedic norms and was a useful rule for settling disputes at the start of the global crowdsourced project. Yet it creates barriers for those consistently underrepresented in writing or publishing: historically marginalized communities and knowledge systems relying on oral traditions.
What is published and where is often predicated on privilege: as we have estimated, only about 7 percent of the world’s seven thousand languages are captured in published material (Graham and Sengupta 2017; Mark Graham is professor of internet geography at Oxford and an advisor for Whose Knowledge?). Ninety percent of all social science and science journals indexed on Scopus and JSTOR are published in English (Albarillo 2014; Glänzl et al. 2019). Additionally, the stories and scholarship of marginalized communities have rarely been acknowledged on par with these privileged sources and, in many cases, are actively undermined.
These kinds of epistemological stances alienate non-white, non-male, non-Western peoples and the knowledge they produce. For example, if the Dalits are considered “untouchable,” testimonial injustice prevents them from challenging this representation and having a voice in how their own history is recorded. This affects how they are represented online based on “reliable” sources (i.e., those that are “verifiable”). The backlash to bringing their knowledge online can feel like a process of retraumatization if we, as allies, cannot support the process effectively.
This effort toward emancipatory, epistemic justice is already underway in the communities we have supported most closely over the past few years and has been described elsewhere (Causevic et al. 2020). Together, we created a collaboratively written set of resources called “Our Stories, Our Knowledges” (Whose Knowledge?, et al 2018), detailing the specific, located work by each community on Wikipedia and the broader internet. We share three such stories, beginning with the story of the Dalit community.
Debrahminizing as decolonizing on Wikipedia
We partnered with Equality Labs, a Dalit feminist collective based in India and the United States that has been organizing Dalit History Month for several years, with an aim to share the contributions to history from Dalits around the world. Dalits are the community of more than two hundred million people, formerly and pejoratively known as “untouchables” across India and South Asia, who continue to be oppressed by a caste system that believes they are reduced by birth to being manual scavengers and sanitation workers. In particular, members of the Brahmin or priest caste in the system were traditionally considered the keepers of scriptural “knowledge” in particular and education in general, and this learning was explicitly and brutally denied to Dalits and others “lower” in the caste system. “Debrahminizing” for Dalit scholars and activists, therefore, is their specific form of resistance and decolonization. Organizers from Equality Labs and the broader community had already compiled a vast amount of content in 2016 for a timeline of Dalit history (Dalit History Month Timeline, n.d.), as part of their effort to debrahminize South Asian history, and they wanted to see more of this history accurately reflected on Wikipedia.
Whose Knowledge? supported Dalit organizers from Equality Labs as they generated a list of Dalit people, issues, and events that were either missing or misrepresented on Wikipedia, and compiled sources for this knowledge. We helped these organizers set up a series of Wikipedia editathons (social events to improve Wikipedia) where Dalit and allied participants in India and the United States met in person to create and improve nearly two hundred Wikipedia articles. For each event, Dalit organizers led outreach in their community, set up and facilitated the physical space, and taught newcomers how to get started.
Even as we built trust and worked together, the Dalit community was being targeted by upper-caste fascists in India and in the United States, and the organizers we worked with continue to be harassed and have to fight offline battles outside of our shared work. Our timelines, therefore, have had to be very flexible, with security always coming first. Dalit contributions to Wikipedia, too, continue to be contested. After the 2017 editathons, a longtime Wikipedian made hundreds of edits to remove Dalit contributions, promoting mainstream upper-caste or Western narratives as “neutral” while erasing Dalit perspectives and sources as “biased.” In response, we changed our strategies: we decided to focus on fewer articles but on those most critical to Dalit knowledge, and we asked longtime Wikipedian editors to support the new Dalit editors beyond the editathon events in fighting this backlash to their work.
Queering as decolonizing on Wikipedia
Similarly, we worked with Okvir, a feminist LGBTQI+ group in Bosnia and Herzegovina that has been collecting digital histories for an online queer archive (“Kvir Arhiv” 2017), beginning with stories from activists and others about their experiences during the war in the 1990s.
Okvir built a list of people and topics they wanted to add to Wikipedia and encountered significant homophobic backlash while working on a handful of biographies of queer activists. The organizers also generated a timeline of queer history, using open-source tools and borrowing ideas from the Dalit History Month example. As Okvir continued to collect video and audio materials for their online archive, we wanted to support this community to ensure their oral histories would continue to be visible and accessible online. We served as a connector to partners like the Internet Archive that can provide robust storage and access, as well as tools for contributors around the world to add new materials to the collection over time.
Unsettling as decolonizing on Wikipedia
The Kumeyaay Nation is a Native American community whose lands stretch across Southern California, in the United States, and Baja California, in Mexico. Kumeyaay scholars and activists have been working to correct an oppressive historical narrative that has been written and circulated for hundreds of years by white colonizers and that is still preserved today in California textbooks as well as Wikipedia.
The Kumeyaay, like many marginalized communities, have been fighting a host of daily battles, and they prioritize community initiatives based on consensus. As a result, workshops and conferences with the Kumeyaay sparked many interesting conversations and projects to explore together, but the community needs to reach consensus on what to do next. Oral history appears to be a more useful entry point with some members of the Kumeyaay Nation, rather than Wikipedia. A Kumeyaay ally, who is herself a Shoshone Indian from the Yomba Band, is a professor of tribal practice at the University of San Diego and has begun piloting rewriting Wikipedia in her “indigenous decolonization” course.
Epistemic or knowledge justice through feminist technologies is not work that can be done alone. Other critical communities and allies that we work with include Wikipedians and Wikimedians, who are the broader movement of volunteers who work on the free and open online knowledge projects (including but not limited to Wikipedia) supported and hosted by the Wikimedia Foundation. We also work with technologists, academics, librarians, archivists, and others, who are helping us create multiple online spaces that can host and offer marginalized knowledge in powerful and accessible ways (Graham and Sengupta 2017).
Devising Postcolonial Approaches for a Decolonized Internet
Audre Lorde (who described herself as “Black, lesbian, mother, warrior, poet”) has famously said: “For the master’s tools will never dismantle the master’s house” (Lorde 1984). She called upon feminists—particularly white, heterosexual feminists—to recognize how racism, patriarchy, and heteronormativity had influenced their notions of feminism, and to acknowledge they often still stood inside “the structures” of power. She continued: the master’s tools “may allow us to temporarily beat him at his own game, but they will never enable us to bring about genuine change.” Crafting new tools, methods, and measures is challenging—as is bringing about genuine change. At Whose Knowledge?, we consider ourselves pragmatic revolutionaries: we recognize we use the master’s tools in certain contexts, as we continue to imagine and plan for transformation.
In this section, we detail three key aspects of our decolonizing approach, created in partnership with the communities we support: in our design, in our processes, and in the ways in which we understand and define “success.”
As we try to articulate a progressive vision of a truly decolonized internet where all forms of knowledge production are valued, we recognize that not even a fraction of that reality can be achieved without a fundamental change in the design of the online and offline platforms that have successfully monopolized production and dissemination of knowledge in today’s world. Our work, therefore, has led us to rethink the design of the sociotechnical systems and processes that will represent the knowledges of the world for the foreseeable future. We design our work, keeping the principles of postcolonial computing in mind.
Postcolonial computing(Irani et al. 2010)—an “alternative sensibility” to design and development—is an approach articulated by a group of researchers in science and technology studies in the United States that includes postcolonial theorists and feminists of color. As they define it, postcolonial computing is “not a project of making better design for ‘other’ cultures or places. It is a project of understanding how all design research and practice is culturally located and power laden, even if considered fairly general” (Irani et al. 2010). These researchers urge us to examine how unequal and uneven power relations are enacted in design practice; they believe this interrogation can lead us to hybrid and culturally dynamic forms of design and technology. Embodying these design principles, another set of women researchers in human computer interaction studies in the United States (Dombrowski, Harmon, and Fox 2016) call for three primary commitments from all designers: to conflict, to reflexivity, and to personal ethics and politics. In particular, the commitment to conflict is critical to the work of knowledge production: we are asked to not depoliticize conflict or paper over it as a design flaw. Instead, we are asked to embrace it in the design process, “partnering with those who have experiences with and/or work to end oppression (e.g., based in class, race, gender, ability, sexual or gender orientation)” (Dombrowski, Harmon, and Fox 2016).
In our work, we design our collaborations with marginalized communities by centering their knowledge and their ways of knowledge production. We design for recognition (Dombrowski, Harmon, and Fox 2016) in two ways: First, the stories of struggle and resilience of marginalized communities like the Kumeyaay or the Dalit are foregrounded in the spaces we create together. Second, we honor the knowledge of these communities as both embodied experience and formal scholarship. In doing so, we believe our convening of activists and scholars from these communities with their allies is a bringing together of multiple forms of expertise; no one community is the “subject” of expertise from others.
We are committed to centering the priorities and needs of the communities we work with through a practice of deep allyship. We approach allyship very much as practice and process: we do not believe we can express the principles of being an ally without embodying the ways in which we engage as allies (“walking the walk,” in other words). This is often complex and messy, and needs significant honesty and generosity from all sides. Decolonizing our processes and methods of working requires an openness to uncertainty and a potential lack of resolution: a certain kind of comfort with discomfort.
In this section, we offer some of the core ways in which we do our work, and examples that embody them.
We meet people where they are, even as we convene and amplify.
We work with a feminist anticolonial framework that localizes and contextualizes the work.
We use open-source and open-culture principles, as we are committed to sharing our work freely, transparently, and as easily as possible.
We balance openness with safety and security. Although we support open-source and open-culture work, we are not evangelists for it above all else, because the safety of marginalized communities needs to come first.
The language of the internet and research about the internet is often imbued with Western assumptions and norms (Smith 1999). So, too, is the language of evaluation. We aim to bridge this gap in language and evaluation by reconciling what James C. Scott (a white American scholar at Yale whose work focuses on state formation and agrarian societies) refers to as “the vernacular” and “the official.” We recognize the values and assumptions inherent in the language we use (Scott 2012). Essentially, decolonizing our metrics begins with decolonizing ourselves. We have learned to recognize “success” as led by the outcomes our communities seek, not only those we consider appropriate or intended. Practically, this means that Whose Knowledge? adapts to meet the needs and demands of different contexts, communities, and cultures — often using approaches and tools that researchers and grant makers in the Global North may not always acknowledge. As allies, we ask: what are the goals of the community we’re working with, and how will they know they’ve achieved them?
Program evaluation tends to prioritize quantitative measures to determine effectiveness and success—constructs that are inextricably tied to a positivist stance: the things that matter can be counted. However, Whose Knowledge? has found a way forward inspired by a 2011 report on “the metrics of movement building” (Pastor, Ito, and Rosner 2011, 1). For any movement-building activity like organizing, research, and communications, the authors emphasize the importance of accounting for transformation. While transactional metrics (what can be counted; e.g., number of people, amount of content created) are necessary and helpful, we also imagine and emphasize the development of transformational metrics (what is critical but intangible; e.g., relationships and collaborations established, social or cultural change).
Concluding Thoughts: Decolonization Is a Work in Progress
Decolonization is always a work in process: within ourselves, and within the organizational spaces we convene. This is particularly true when we think about the internet. Despite the history and knowledge of marginalized communities being the history and knowledge of the majority of the world, the internet does not reflect that reality. It continues to be designed by a minority, without centering the needs and aspirations of the majority. In our work to transform the nature and form of online public knowledge, Whose Knowledge? attempts to embody and practice decolonizing design, process, and metrics.
We continue to learn from the extraordinary courage and resilience of the communities we support, and are inspired by them and others daily. We also honor deep allyship: the ways in which individuals and organizations can meaningfully support marginalized communities by considering allyship as an act of convening multiple forms of expertise and experience, centered in the community.
Most significantly, we continue to be driven by the feminist, collective inspiration of what our communities believe about knowledge justice: “Our knowledges are urgent. They are practical. They are creative, colourful and collective. They are plural. That is why for us, ‘knowledge’ is never singular. Our knowledges are transformative. They are hope.” (Whose Knowledge?, et al 2018, 22).
Camille Emefa Acey is the global head for customer experience at Humio, a logging and aggregation platform. Based in the United States and Slovenia, she is a founding member of the Collective for Liberation, Ecology, and Technology, a radical feminist tech collective in Brooklyn, New York.
Siko Bouterse is cofounder and outgoing codirector of Whose Knowledge?, starting in January 2021. Based in California, she is a community organizer, facilitator, strategist, and sometimes grant maker, with significant experience working with international online communities.
Sucheta Ghoshal is an incoming assistant professor in the Department of Human Centered Design & Engineering at the University of Washington, beginning in January 2021. She has been embedded in grassroots social movements in the United States—both as a researcher and as an activist—for the last three years.
Amanda Menking is an associate teaching professor at the Information School and the director of the Master of Human-Computer Interaction and Design Program, both at the University of Washington in Seattle. As a researcher, she is interested in knowledge production in online communities.
Anasuya Sengupta is cofounder and codirector of Whose Knowledge?. From India, and now in the United Kingdom via the San Francisco Bay Area, she has led initiatives in India and the United States, across the Global South, and internationally to amplify marginalized voices in virtual and physical worlds.
Adele Godoy Vrana is cofounder and codirector of Whose Knowledge?. Now based in California, Godoy Vrana has led business development and partnerships initiatives to help build more plural and diverse communities in her native country of Brazil and globally.
The many individuals, organizations, communities, and movements we work with and learn from, every day. You inspire us.