For more than 15 years, Wikipedia discussed what to call the third child of Ernest Hemingway, a doctor who was born and wrote books as Gregory, later lived as Gloria after undergoing gender-affirming surgery, and, when arrested for public disorderliness late in life, used a third name, Vanessa. Last year, editors on the site finally settled the question: The Gregory Hemingway article was deleted, and its contents were moved to a new one for Gloria Hemingway. This would be her name going forward, and she/her would be her pronouns.
Wikipedia’s billions of facts, rendered as dry prose in millions of articles, help us understand the world. They are largely the brain behind Siri and Alexa. They have been integrated as official fact-checks on conspiracy-theory YouTube videos. They helped train ChatGPT. So, unsurprisingly, when you search Google for “Gregory Hemingway,” it follows Wikipedia’s lead: You are told about Gloria instead.
In Wikipedia’s early days, the question of what to call Gloria Hemingway would have been treated as a quick mission to locate a fact in established publications such as The New York Times. Joseph Reagle, a Wikipedia expert at Northeastern University, told me the site has an inherent “conservatism,” faithfully reporting whatever secondary sources say about a subject. And at the time of Hemingway’s death, in 2001, no major publication, including the Times, called her Gloria.
But in recent years, something has begun to change. Wikipedia’s editors are no longer simply citing dated sources; instead, they are hashing out how someone would want to be understood. But even though these deliberations touch on some of the most controversial issues around—and reach conclusions that reverberate far beyond Wikipedia’s pages—they are shockingly civil and thoughtful for the internet today.
The breakthrough idea of Wikipedia was supposed to be its biggest vulnerability. “The encyclopedia anyone can edit” threw open the gates to whoever had something to contribute, turning Wikipedia into one of the most visited websites on the internet. But who was to trust something “anyone” may have written? The site definitely has inaccuracies; any student working on a research project has gotten a spiel about how Wikipedia will lead them astray.
Of course, only a tiny percentage of Wikipedia’s visitors actually take up the offer to contribute. There are campaigns to draw in new editors, especially given that the existing ones skew heavily white and male, but the most reliable motivation for getting involved seems to be the urge to fix something wrong as opposed to create something new. Articles typically start off small and stubby, perhaps even inaccurate, and are steadily improved and corrected.
The desire to fix something wrong—in this case, articles that have not kept up with the times—is meant to play out on an article’s “Talk page,” a companion page dedicated to discussing edits. Take the debate over Gregory versus Gloria. Last February, Hemingway’s Talk page fielded a proposal on what name to use. There was a week of debate, long discussions in which a dozen or so editors grappled with how Hemingway would have wanted to be perceived. The main advocate for moving the page from Gregory to Gloria was an editor named TheTranarchist, and the main opponent was an editor named StAnselm, a self-described Calvinist who has created more than 50 articles about biblical characters and scenes. Yet the discussion on the Talk page was about facts and Wikipedia policies and guidance, not politics. “It didn’t seem culture warrior–ish,” Reagle said.
The discussion ended with a hung jury: seven editors for Gloria, seven for Gregory. An experienced editor, Sceptre, stepped in and ordered the article to be renamed. The decision was appealed, and an administrator concluded that Sceptre had made a tough call that was ultimately reasonable. On the biggest social-media sites, such a decision might have descended into endless mudslinging. Instead, everyone has respected the outcome and moved on. The article hasn’t been touched in five months.
Exactly how these deliberations play out is different from article to article, but what’s changed is that Wikipedia is no longer automatically outsourcing the decision to a judgment of the past. The point isn’t that Wikipedia has gone “woke.” Sometimes the deliberations don’t lead to any fundamental changes at all.
That has been the case with the page for the late pioneering legal scholar and Episcopal priest Pauli Murray, which has periodically ignited pronoun fights from readers who want to right what they see as a wrong. Murray used she and her in her own writings but, in today’s terms, might have been considered nonbinary or a trans man. As one conflicted editor wrote on the Talk page, “If Murray were alive today, Murray would probably use he/him/his or they/them/their pronouns. The question is do we have a right, or an obligation, to apply these retroactively? Is it okay to be anachronistic in this matter? I do not have answers to these questions, which is why I am calling attention to this.” Wikipedia’s editors have begun grappling with tough, even existential questions that might have traditionally been the domain of historians rather than encyclopedias.
There has been a similar attempt to interrogate understandings of the past by renaming the articles about a series of places whose names contain squaw, including the California valley where the 1960 Winter Olympics were held. On occasion, editors would propose such a move, noting that squaw is considered a slur against Native Americans. Others would say that as an encyclopedia meant to be helpful to people, Wikipedia should use the most common name. “The Olympic Games of Squaw Valley” are embalmed in the past, they argued, so how can the name “Squaw Valley” be removed?
In September, when the federal government said it would begin the process of officially scrubbing squaw from place names, a proposal to rename the article about the California valley succeeded. Case closed. But take a look at the Talk page, and you’ll find a level of discussion that more resembles the collegiality of a workplace than a network of unpaid online commentators. The experienced editor who concluded that the community favored renaming the article confessed that he had been a bit confused by the issue. “Forgive me,” he wrote, “but just as I fail to understand other forms of ethnic slur, I am hard-pressed to make out why Native Americans would consider the naming of anything, a valley, a town, a waterfall, anything, after the general term for ‘spouse’ would be indigestible. If it were called ‘Spouse Valley’ or ‘Wife Valley’ I don’t think any ethnic slur would be sensed by anybody … Would really appreciate any light that is shed on this subject!”
Wikipedia has long represented a fundamentally unique form of information production—it isn’t credentials based, or top-down like Britannica. That’s not to say that it’s perfect; the site has all the secret hierarchies, obscure rules, and confusion we’d expect. At times, it has been a vector of misinformation. But as the site takes on thornier edits, what it means to be a Wikipedia editor is changing too. By wading into factual dilemmas instead of deferring to secondary sources, editors have assumed a new level of authority. The results will be choppy and contradictory; proposals for tweaks will come from ordinary readers and editors who have been moved by offense, and questions will be decided through deliberation, often with great self-seriousness.
After all, these small decisions do have real consequences. Wikipedia results spread across the internet, often influencing what we think of as reality. “I don’t think any community project has as much reuse and significance for the rest of the world that Wikipedia does,” Reagle said. Indeed, Google “Squaw Valley,” and you don’t see the term at the very top. Google does, however, suggest the question “Does Squaw Valley still exist?,” which it answers with a Wikipedia excerpt explaining that it remains but that the name has been changed “due to the derogatory connotations of the word ‘squaw.’”