As you’re reading this, chances are you have consumed some content from Wikipedia today. It powers the search results for Alexa and Siri. It has articles on everything from Snooker to the Vice President on the tv show “The West Wing”. It is the 5th most popular website in the world, and the only one run by a nonprofit. It is the largest encyclopedia in human history, one of the largest repositories of knowledge ever compiled, freely available, studiously accurate, and fueled by anonymous online commenters. It’s a modern miracle.
That’s not to say it doesn’t have flaws. Speaking as a longtime editor, it has plenty. And there are a lot of active debates about editorial decisions within it. But through a lot of hard work, we have mostly rendered those shortcomings irrelevant today. When you read a Wikipedia article, you are probably getting unbiased, factual information that would pass the sniff test of an expert in the field.
None of that was guaranteed from the outset. In fact, in 2004 or 2005 the reliability we now consider the base standard would be almost unimaginable. Like the primordial Earth, we went through a turbulent period of chaos and flame in our first years. From those vicious forum battles emerged consensus policies on the conduct of our editors and what could go into our articles. Those policies, and our devotion to upholding them, has protected us from the worst impulses of online discussion and made Wikipedia what it was today.
I wasn’t active in the community at the time: I was in the fourth grade. But the debates are still available, if you know where to look. And I can see the impact these policies have had, not only on the stellar quality of Wikipedia today, but through a case study in its opposition. Conservapedia.
You could be forgiven for never having heard of Conservapedia. Since it was created over a decade ago, it has gotten over 700 million views, slightly more than Wikipedia gets per day. But it is still our Doppelgänger, a reminder of what we could have been.
First, some history. Conservapedia was created in 2006 by a man named Andrew Schlafly. He was the child of Phyllis Schlafly, a conservative activist best known for helping sink the Equal Rights Amendment in the 1970s. His educational pedigree is impressive, going from Princeton to Harvard Law before becoming a conservative activist. He started editing Wikipedia in 2005, primarily focusing on his mother’s article and topics related to the pseudoscientific theory of intelligent design.
Note: a good metric for how contentious a Wikipedia article has been within the editor pool is the length of its talk page. Our article on Intelligent Design has the longest talk page on the whole goddamn site.
These edits had a clear point of view. Andrew Schlafly felt Wikipedia was being unfair to the Intelligent Design movement. As a result, he found very little traction, and in 2006, he split off and founded Conservapedia.
I will be generous to its content and say that it is at times not fully accurate. But it serves as an excellent case study, because it fundamentally diverges from Wikipedia in its interpretations of our three core content policies: No Original Research, Verifiability, and Neutral Point of View.
Before I begin, I want to make two notes. First, wiki articles are prone to change, so linking to them to prove a point is risky business. To avoid that, when I cite a Wikipedia or Conservapedia article, I will be linking to the version that was current as I was writing. That way, there’s no updating the articles after-the-fact.
Second, Conservapedia has a far smaller editor base than we do: as I write this, there are about 150 regular editors on the site. With fewer eyes, it’s harder to maintain the same kind of stylistic professionalism Wikipedia upholds. Out of respect to those editors, who are doing their job within the confines of Conservapedia’s mission, I’m not going to be discussing any spelling, grammar, or formatting errors. This site is meant to be a serious alternative to Wikipedia for people who feel we have failed to properly adhere to the facts. We’re going to treat them seriously.
One of the problems Wikipedia faced early on is that our random anonymous editors may not be remotely qualified to talk about the topic they’re discussing. For instance, while I’m a pretty smart guy, I have only a limited education in political science. But I’ve added content to over a hundred articles in the topic. You don’t know who I am. You don’t know what my credentials are. Why should you trust what I write?
For that reason, we ban original research. Let’s say you’re a mathematician, and you discover a proof of some unsolved problem. You can’t then put that proof on the Wikipedia article for that problem, even though it is entirely mathematically correct. It’s original research: stuff you put together and found out yourself.
That changes once it gets published in a journal. We are scrupulous about our sourcing, and want every statement to be cited to an expert on the subject. Unlike most other outfits, we don’t prioritize primary sources. While they’re allowed, relying on them means we’re relying on our editors to interpret them, and find the most pertinent accounts from their own expertise. We prefer to outsource that and rely on secondary sources, analysis of those accounts by experts. That way, you don’t have to trust us.
Conservapedia explicitly does not share this policy. Scan through their list of how their policies differ from ours, and you will find this:
“We allow original, properly labeled works, while Wikipedia does not. This promotes a more intellectual atmosphere on Conservapedia. On Wikipedia, observations based on personal experience and interviews have been dismissed as “original research.” Here, we do not restrict research for articles in that manner.”
It is an understandable decision on their part. If you feel the reliable sources we use (mostly journalists) carry an unrecognized political bias that leads them to not be factual, you might want to do your own interviews. See also, “We do not allow opinions of journalists to be repeated here as though they are facts. Instead, we require authoritative support.”
The result of that decision, however, is that you give the opinions of your editors an undue level of weight. Scroll through their citations, and you start to find blogs, even on their most popular articles. To be fair, these aren’t just personal blogs from the writers. Mostly. Even with that caveat, however, none of these would qualify as reliable sources on Wikipedia. They might be accurate. They might be groundbreaking and innovative arguments that will be uncontroversial in a few short years. But we won’t cite them.
They also are fans of linking directly to YouTube videos. On Wikipedia, we prefer not to simply throw clips at people. Especially when they can be quite long and not searchable, we prefer to provide context. More notably, however, it would be original research. We allow the use of primary sources to cover only the most uncontroversial factual claims. For instance, you could use quotes from a novel to summarize the plot. Any interpretation of those sources, however, must itself be cited to another secondary source. We also do not allow what we call synthesis: basically, if I find an article that says Mike Pence was photographed buying a strap-on at his local sex shop, and an article saying men buy strap ons so their female partners can peg them, I still can’t use those to say Mike Pence secretly loves being pegged, and not just because no one wants to think about that. You can’t connect multiple reliable sources together to say something none of them individually said.
Conservapedia does this constantly, in a series of articles that have no counterpart on Wikipedia. I call them the “arguments” articles, and they are generally a list of points for, or refutations of points against, some topic in science or politics. The worst offender is probably that on Obama’s Religion, but you can also see it in Counterexamples to an Old Earth. I’ve linked you to a particularly clear example. Here, they cite a piece on declining SAT scores and a number of pieces of historical discourse to demonstrate that our intelligence has been going down over the last few centuries. None of their sources individually discuss this. That is the analysis of whoever wrote that bit of the article. Schlafly himself, in this case.
Now, there are drawbacks to Wikipedia’s approach. Our dedication to the secondary sources means that if they get it wrong, we get it wrong. Or, in the words of one of my favorite essays on Wikipedia policy, “If Wikipedia had been available around the sixth century BC, it would have reported the view that the Earth is flat as a fact without qualification. It would have also reported the views of Eratosthenes (who correctly determined the Earth’s circumference in 240 BC) either as controversial or a fringe view.”
But that doesn’t make Conservapedia’s approach the right one. In this case, it’s led them wrong. While IQs have been decreasing recently in Europe and the US, this is a fairly recent phenomenon. And if they had consulted a few more experts before citing the civil-war letters and the Lincoln-Douglas debates, they might have learned that the 19th century was just as full of inane babble as today. After a hundred years, the babble gets forgotten, and the gems get preserved. The supposed greater intelligence of the past is an optical illusion. However commendable their interest in independently researching their subject matter, it’s caused them to factually report claims that are fundamentally wrong.
The second core content policy is pretty self-explanatory. As a rule, you should be able to independently verify any statement you read on wikipedia, solely by following our footnotes. We do not half-ass this: as I write this, the article on Donald Trump has 808 inline citations. While most articles aren’t that excessive, even fairly obscure articles like that of the Jurassic fish Leedsichthys will have a citation every hundred words.
We invest a significant amount of effort into making sure those references stay valid. When the url we’re citing ascends to the great firewall in the sky, we use the Internet Archive’s Wayback Machine to resurrect it. We’ve even created bots that automate the process, and when they see sources that are still alive but don’t have an archive backup, they’ll make one. Just in case.
We also do not, under any circumstances, allow circular citations: Wikipedia articles that cite other Wikipedia articles. Our articles are always subject to change as we learn more, so a citation claiming that we said something has no guarantee to still be a true citation in the future. That seems simple, but remember, there are a lot of outlets in the world that will just copy/paste content from our articles and report them as fact, and a lot of forks of our content to provide more reliability to other people.
XKCD fans will be familiar with the concept of citogenesis. That comic explains it better than I could, so just check it out. This is something we worry about a lot, and it’s happened several times already. Other editors’ experiences may vary, but far more of my time is spent verifying sources or looking for new ones than actually writing copy for any of our articles.
To be clear, we aren’t perfect: particularly in obscure entries that are only rarely read, things can slip through the cracks. And because we’re so beholden to the reliable sources we can find, our content gets biased towards subjects where information is available online. But that doesn’t change our general dedication to the principle, and efforts to apply it as best we can.
Conservapedia agrees with this idea in principle. In fact, verifiability is their first commandment. However, it adds a caveat through its fourteenth difference between it and Wikipedia.
“We do not require contributing editors to have to explain themselves constantly and justify every single edit to prove that it conforms to an exacting set of rules which are designed to suppress original thought, new ideas and penetrating insights.”
I sympathize with this sentiment: the many rules and guidelines for Wikipedia’s content make for good articles but a high barrier for entry, and it’s too easy to bully new editors with a combination of aggressive policy-citing and faux-legalese. But in practice, their solution makes for some pretty sparse references sections. As I write this, their article on the benefits of Capitalism has ZERO citations or footnotes, and has never had one for the entirety of its almost nine-year existence. This article is currently the second entry on their main page’s “popular articles” list.
It is far from the only example. There are a host of stubs, mostly denigrating liberals, which have no sources at all. It’s not just Liberal hyperbole, there’s Liberal dislikes, Godless liberal, and Liberal hedonism, which apparently consists of Harlem Shakes and Miley Cyrus.
To be fair, these appear to be more like dictionary definitions, which don’t usually cite sources. But there is also an abundance of articles with nowhere near the amount of sourcing I would expect on Wikipedia. Their article on Theodicy, an important branch of religious studies, has only 2 citations. The Wikipedia equivalent has 110. Granted, our article is a lot longer and more in-depth (Conservapedia explicitly emphasizes brevity over thoroughness), but it’s still 6 times more citations per word. Their article on God has only 4 citations, equal to the number of citations in the first paragraph on Wikipedia.
They also don’t share our concern with circular citations. In their article on liberals, the reference that liberals support “Hatred” links to another Conservapedia article.
This is not just comparing bibliography sizes here. When you leave the references this sparse and discourage double-checking of uncited additions, things slip through the cracks. Take the Benefits of capitalism article again. That last point, the 10th, is complete balderdash. There are countless examples throughout history of Capitalist nations imposing Tariffs, restricting trade with rivals, and more. Free trade is a common aspect of Capitalist nations today, but as their own article on Donald Trump’s economic policy will tell you, they aren’t necessarily the same.
Circular citations, unsupported statements of fact, articles completely lacking in references, these are the side-effects of this relaxation of Verifiability standards. Their decision to allow these gaps may have made their site more welcoming to new editors, but it has resulted in considerably less reliable and consistent articles.
Neutral Point of View
Before we start shamelessly mocking Conservapedia’s political bias, it’s worth taking a moment to think about what a Neutral Point of View means in an encyclopedia. After all, if one person says pi equals 3 and another says it’s an irrational number roughly equal to 3.14, does a neutral point of view mean not saying one of those is right and the other is wrong?
Instinctively, you want to say that neutrality doesn’t apply there because the value of pi is a fact and not subject for debate (note: even that is way more complicated than you’d expect) . It’s an appealing argument. But there are a lot of reasons it doesn’t work in practice. Countless issues today, from Global Warming to the accusations against Brett Kavanaugh, stray into questions of what is factual and what is only “likely” or even “speculated”. There may be far stronger cases for one side than the other, and there may even be a consensus, but often there are detractors who historically have been right on occasion. How do you weigh those competing points of view?
We use the Principle of Due Weight. Essentially, we look at what the reliable sources say, in general, on the topic, and allocate space in the article to all the views they take, proportional to how widespread that view is. About 97% of Scientists agree that Global Warming is a real and human-caused phenomenon, so about 97% of the article on Global Warming is dedicated to that view. Well, sort of. It’s more complicated in practice but that’s our rule of thumb.
Again, we are not perfect in this regard. Wikipedia’s content still subtly reflects the perspectives of those who edit it, though usually more so in what we decide is notable enough to get an article, not the content of the article itself. We also are yoked to the consensus of reliable sources. Remember the flat-earth quote I had above. When the fringe views are actually right, we can’t reflect that.
Conservapedia sees those drawbacks as unacceptable failings. In their own words: “We do not attempt to be neutral to all points of view. We are neutral to the facts. If a group is a terrorist group, then we use the label “terrorist” but Wikipedia will use the “neutral” term “militant”.”
Already, I see real issues with their stance. The line between “terrorist” and “freedom fighter” is famously murky, and there is ample debate even within American political discourse about which groups count and which don’t. There is a reason Wikipedia goes with “militant” and then enumerates exactly which organizations label which militants as terrorist groups. Take the Muslim Brotherhood, for instance. As I write this, Conservapedia includes it (along with both the Ku Klux Klan and John Brown’s Anti-Klan Committee) as an example of a radical leftist terrorist group. That contradicts the US Government’s position on the group under the Trump, Obama, and Bush Administrations. While Russia and several Arab states classify it as such, most of the Western World views it as a legitimate political party, one which elected the president of Egypt for multiple years. It’s an Islamist group, and I disagree with its positions on most issues, but it’s hardly a terrorist organization.
Conservapedia doesn’t just think other ways, it says so as a matter of fact. The Muslim Brotherhood are terrorists. Ipso facto.
A list of all the articles in which Conservapedia abandons a neutral point of view to push a conservative agenda would be a list of all articles on Conservapedia. They keep a running list of everything horrible liberals apparently believe in, and feature it so prominently it’s above the table of contents. Their article on Fake News explicitly defines it as a liberal-only phenomenon in the first sentence. FDR and the New Deal apparently prolonged the Great Depression, referring to a consensus of historians without citing one. They make no attempt to hide a clear political bias anywhere in their articles.
And to be clear, that’s not inherently wrong. If Wikipedia were to suddenly start erasing articles on the atrocities of Stalinism, I would want a more conservative alternative I could ride my flying pig to when I needed information.
It becomes wrong when that strong editorial position spreads ahistorical bullshit all over the site. Their article on William Shakespeare describes him as “anti-feminist”, as evidenced by the content of “The Taming of the Shrew”. This is an impressive trick on Shakespeare’s part, since feminism wouldn’t exist for another 2 centuries after he died. Alexander Hamilton is also a conservative now, which is at least slightly more defensible than Shakespeare. It is still, however, nonsense. He was a federalist with nationalist tendencies, and does not fit neatly onto the left-right political spectrum of 21st century America. Just like every other great thinker who was born, lived, and died hundreds of years before any of us were born.
Nor are they limited to revisionism of the distant past. They describe George Soros, a Hungarian Jew who barely survived Nazi Occupation, as an anti-semite who worked for the Nazis. This particular conspiracy theory is an Alex Jones special which has been boosted by everyone from Glenn Beck to authoritarian president Viktor Orbán of Hungary. It is virulently antisemitic itself, and is part of far-right efforts to blame The Holocaust on Jewish collaborators.
It is discussed without context in the introduction to his article.
Still more harmful is their firm advocacy of conversion therapy. They describe it as an effective and charitable course to correct a mental illness, citing a 2007 study with crippling methodological flaws over the overwhelming consensus of the community of psychologists to claim that it is something other than a hateful and traumatizing form of pseudoscience.
And then there is Global Warming. It is by far the most important issue facing us today. The future of the human race rests on what we do about it in the next 20 years. Yet Conservapedia would have you believe that the science is inconclusive at best and against it at worst. I won’t pretend that this exercise in Young-Earth-Creationist navel-gazing is the source of the right’s climate change denialism. Yet it is still a small contributing factor to the most damaging aspect of the right’s platform today.
The danger of giving full room to fringe theories because the consensus might be wrong is that you are now subject to your most conspiratorial writers. However much I might sympathize with their desire for a Wikipedia that would recognize Galileo in his own time, it has caused their content to reflect the worst of the far-right’s paranoia and hatred.
Conclusion: Why Any of This Matters
It may seem like nothing is at stake here. After all, Conservapedia is extremely fringe even among the Christian right. Steve Bannon and Richard Spencer may believe some hateful shit but even they don’t think the Earth is 6000 years old. It has barely a hundred regular editors, and shows clear signs of minimal activity over the last several years. Remember that Liberal Hedonism article, that talked about Miley Cyrus twerking and the Harlem Shake? There’s a reason it uses examples from mid-2013: It hasn’t been touched by a human being in the last five years.
It’s not completely irrelevant today. It got roughly 50 million more views in the last year, over 100,000 per day. It’s an endless fount of material for people like me to mock, and I’ve barely scratched the surface there. But I see something deeper in this tragedy of an encyclopedia.
Wikipedia’s content and community have survived the worst ravages of the internet, everything from Russian attacks to Gamergate, because our community of editors remains dedicated to a shared set of basic principles. We can disagree on content, style, weight, and more, but we are still working for the same goal. We are not a forum. We are not a newspaper. We do not have an editorial section. We are an encyclopedia. And every participant is onboard with that shared vision.
Conservapedia is a case study in how that vision dies. It began when one editor abandoned the site to create a parallel version that would not contradict what he knew to be right, even when the reliable sources said he was wrong. I know what he felt when he did it. I feel the same way when the decisions those policies make go against my morals. But I didn’t make Shankipedia, nor did any of the countless other editors who have disagreed with on-Wiki consensus, because we remained committed towards the first principles of the site.
However virtuous his motivations may have been, the result is an abomination. What was meant to be an alternative to Wikipedia has collapsed under the weight of its own bias. It is unreliable and inaccurate. It promotes hateful conspiracy theories without qualification. What little is left of its community is toxic and dictatorial towards anyone who disagrees with even minor aspects of their ideology. They are, in short, exactly what you’d expect if you trusted the writing of an encyclopedia to an online forum.
None of that was unpredictable. It is simply what happens when the mods prioritize their personal agenda over what is best for the community.
The story of Conservapedia is a cautionary tale. It is a reminder that platitudes like “Neutral Point of View” can carry a lot of weight when we all believe in them. It is proof that these core principles can work miracles, even on the most fraught and contentious topics. And it is a live demonstration of what happens when we abandon those core principles as soon as they prove inconvenient to us.