The Women that “Category-Gate” Erased

N.B. – this piece was written as an Op Ed in May. After unsuccessfully making the rounds of several major outlets, I’m publishing it here before it becomes too stale. As my first OpEd effort, this piece owes debts to Adrianne Wadewitz, whose work on this same topic inspired these thoughts, to Alex Juhasz and Jessie Daniels, who helped with drafting and editing, and Amy Guth and Audrey Bilger, who both offered mentoring on the pitching process. The process has been a great example of the ways that feminist networks can help support new work by women.

—-

Amanda Filipacchi’s April 24 New York Times piece opened up a public debate about Wikipedia practices and drew attention to the categorical and structural sexism made evident by John Pack Lambert’s editorial choice to create and use an “American Women Novelist” subcategory. Several pieces followed in the NYT, on Salon.com, Women’s News.org, NY Books, The Atlantic online, Motherboard, and an array of personal and professional blogs. While internal conversations about bias, including gender bias, date back at least to the 2004 launch of the WikiProject “Countering Systemic Bias,” Filipacchi’s piece and many of those that followed suggested that Wikipedia unthinkingly reflected a “universal bias” and that male editors were ushering us all in taking a “step backwards” in terms of gender equity.

Dubbed “category-gate,” the dust up has helpfully drawn attention to issues around gender, technology, and how we make knowledge. Unfortunately, Filipacchi and others generally ignored the ongoing work of women and feminists to address bias in Wikipedia’s content and structure, perpetuating the very kind of sexism they purported to unveil.

Wikipedia is a go-to destination for information and it is an ever-evolving resource of impressive size. A look today calculates the English language site with 30,119, 512 pages and the growth chart for articles is a steep curve. Use has also expanded; 476 million unique viewers have visited Wikimedia sites. Even professors, who once banned the use of Wikipedia in classes, make use of the resource. The English Wikipedia site is ranked as the sixth most popular site by Alexa, a web traffic and data analytics company. User demographics skew young, male, and affluent and for these users Wikipedia is a major source of knowledge.

What few people realize, even so-called “digital natives” who’ve grown up with the Internet, is that Wikipedia is authored by the same pool of people who most regularly use it; while anyone can edit Wikipedia, the majority of editors are young, white males. But this isn’t common knowledge. Jessie Daniels, Professor of Sociology at CUNY, working with teenagers (15-19 year olds), found that most of them didn’t realize they could edit Wikipedia entries. Nor did many of these users recognize that there isn’t a “Wikipedia” company to blame or praise.

Wikipedia is a community developed knowledge tool. Editors are volunteers who work together to collaboratively create the world’s largest knowledge resource.  While there is great democratic possibility for “anyone” to edit, the truth is that there is a relatively small percentage of total users who do. That’s not surprising really. It’s how most crowd-sourcing works, and in fact, how most volunteer organizations run  – your kids’ school bake sale or the food co-op.

Notably for “category-gate,” the volunteers at Wikipedia are mostly young men. This exposé noted a visible symptom but failed to take account of the fact that Wikipedia is community of editors self-governing under “five pillars”: a set of community policies that govern both the nature of acceptable content and community behavior. In addition to editing and researching content, the work of Wikipedians also includes participating in the discussions on the talk pages of categories, and self-governing in relation to community standards of behavior.

The point here is not that the self-governance absolves the encyclopedia’s structural sexism, but that we cannot possibly address the systemic bias of a resource that is fundamentally misunderstood.  People were up in arms about Lambert’s choice to create a subcategory of women writers and to move women out of the general American novelist category, thereby suggesting that only men are “American Novelists.” And rightly so.

But the response of Wikipedians was different; they discussed this controversial issue as they do all others on the site. The Talk pages for editors (which are easily viewable for day-to-day users, too) quickly filled with debate about the organization of the list of American novelists and of one editor’s choice to subcategorize by gender. The community ultimately chose a gender-neutral solution with “almost unanimous agreement” according to Adrianne Wadewitz, a Wikipeaida Ambassador who is also a feminist scholar of English Literature with a post-doc in New Media Studies at Occidental College.

Filipacchi’s article prompted Wikipedians to take a closer look at Lambert’s work and to discuss the issues. It also drew brief public attention to the craftedness, the incompleteness, and the collaborative nature of Wikipedia. Unfortunately, that attention was then redirected to the work of an editor known as “Qworty,” as juicy examples of what Andrew Leonard of Slate.com has dubbed the “age of revenge editing.” Leonard’s story drew readers’ attention to a sexier narrative: a story about rampant sexism in what is often mistaken for a corporate product, a digital product (mis)shaping how we understand the venerable novel. Such coverage riles up readers; it encourages shares, likes, and a lot of hand wringing. It was also an opportunity lost.

While communal agreement isn’t particularly sexy, the work and opinions of feminist Wikipedians is apparently even less so. Not only were reporters uninterested in the nuances of community and crowd-sourced knowledge production, those who contacted Wikipedians seemed to be explicitly uninterested in hearing the voices of female and feminist Wikipedians. As Wadewitz has observed in her own blog-post about the events, articles trumpeting “Wikipedia’s Sexism” or the encyclopedia’s  “Women Problem” took up a preexisting narrative that put sexist men at the center of Wikipedia and women on the outside.

This story excludes the important and longstanding work of women and feminists in the Wikipedia community, including WikiProjects addressing gender disparity in content and systemic bias. Nowhere mentioned were the WikiWomen’s Collaborative or the new Wikipedia Teahouse. Dramatic stories of revenge editing occluded as well the extensive work of the international feminist collaborative known as FemTechNet to initiate and support the training of women and students, many of whom come from underrepresented groups, to edit Wikipedia, and think about women’s relationships to technology, knowledge-production and internet community. Related efforts, like #toofew—another effort to write women into Wikipedia—and WikiStorms held at Feminist That Camps and through the fembot collective—indicate a productive, collaborative feminist effort at a much deeper, and dispersed level, than the questionable and questioned activities of one editor.

Furthermore, the media’s first gloss on the story entirely failed to take account of the complex community of editors and the work they actually do. Such reporting also suggests that addressing systemic bias is a numbers game – add more women and you’ll get equity. Unfortunately, the systemic bias in this case isn’t just about who is doing the editing; it’s about what counts as knowledge.

Wikipedia is an encyclopedia. Editors are creating a “tertiary resource,” meaning that everything that
goes up on Wikipedia does so in the spirit of aggregating accumulated knowledge. First and second person accounts are generally not strong enough support on their own for a claim or change on Wikipedia. Published sources are the gold standard. What’s more, the “notability requirement” means that it’s not enough for an article to be verifiable, it also needs to be important. This means that in addition to worrying about the bias produced by a relatively homogenous editorial community, we should be worrying about the biases expressed in our collective sources of “important” “accumulated knowledge.” Perhaps even more important than the misogynistic work of particular editors, are the biases of print culture that Wikipedia is at risk of reproducing – silently.

Yes, women should be editing Wikipedia in ways that more accurately reflect our numbers in the population as a whole. But this isn’t enough. It’s too easy to focus on John Pack Lambert’s work and politics, to make “Qworty’s” obnoxious revenge quest the easy target of our (justified) anger, to think about the individual actions and not the structures of bias that continue to exclude women, people of color, and queer people from processes of knowledge creation.

Anyone interested in how we come to know what we know should be concerned about the systemic bias of Wikipedia. But they should be equally concerned about media representations that hide more pervasive forms of bias by focusing on individuals. By ignoring the work of women and feminists currently underway, such stories argue that women don’t edit because the environment isn’t friendly enough or they are too busy tending to families, thereby perpetuating a popular sense that technological work is men’s work. Such reporting created the very kinds of gendered discrimination it purported to uncover.

As a feminist digital humanities scholar in the broader network of scholars and feminists concerned with issues of gender and technology, it is my hope that “category-gate” instigates a broader conversation about the manifold ways that women are still seen as a “subcategory.” More importantly, it should encourage us to acknowledge that we are responsible for crafting knowledge and that we all need to understand what Wikipedia is and how it is created. It’s the world’s largest knowledge resource and it’s a place where women and feminists are working every single day to ensure that it represents the world we want to live in.