The War of the Words: How ‘Truth’ is Altered on Wikipedia
We’ve all used Wikipedia. Over its twenty-year existence, the encyclopaedia, run by volunteer editors, has become an essential means of research on almost anything or anyone. Wikipedia has been lauded as a bastion of neutrality in an increasingly polarized internet landscape. The Wikipedia website even declares itself towards ‘representing fairly, proportionately, and, as far as possible, without editorial bias’. Yet, behind the metadata of all its articles lies a problem so crucial as to put not only the project into question, but questions whether a universal bank of knowledge, supposedly accounting for all or as many voices as possible, can exist in the first place.
Although Wikipedia is founded on a rejection of editorial bias, a large number of articles contain unbalanced perspectives or an overlooking of certain, sometimes crucial, viewpoints. Identity, ideology, geography, gender, and other factors can complicate the website’s mission for a neutral encyclopedia. Whether explicit or implicit, falsities and distortions are abundant in some articles, especially those that discuss politics, geography, and culture.
In 2009, Croatian users organized an attempt to alter political pages on the Croatian-language Wiki. A number of ultra-right and conservative users began falsifying articles, promoting fascism and historical denialism while castigating left-leaning viewpoints, the LGBTQ+ population, and inflaming an already tense ethnic conflict. Editors attempting to fix the inaccuracies were routinely harassed and banned. It took over ten years for any substantial action to be taken, when, in 2021, the Wikimedia Foundation, the parent organization of Wikipedia, banned the accounts responsible and reverted their edits.
This takeover was by a relatively small group of individuals on a Wiki magnitudes smaller than the English-language version, but does that mean that the falsehoods are explained merely through a few rogue users in a small community?
Japanese-language Wikipedia, with ten times the articles as the Croatian version, is one of the largest communities on the website. It has never had any major issues with rogue users. Yet, it has been accused of serious revisionism, especially when it comes to the Nanjing Massacre. Florian Schneider studied in 2018 discrepancies in the Japanese article on the Massacre, remarking the article ‘attempted to justify the rape and murder of Chinese civilians’. There was only one image in the entire article, and it didn’t depict any acts of violence. The first paragraph even doubted certain well-known facts, casting the Japanese army as less guilty. A similar story can be said for Turkish Wikipedia’s accounts of the Armenian Genocide.
Even on English-language Wikipedia, political articles, especially those relating to US politics, are slanted. Articles pertaining to healthcare, gun control, and civil rights skewed towards left-leaning language and had a ‘systematic prevalence’ of left-leaning sources. Issues of immigration trended more towards the right and tended to use more conservative sources. One explanation is that Europeans and highly-educated Americans – those who edit US political articles the most – are, on average, more left-leaning than the average American, yet this does not explain the rightward slant on some other articles.
What is understood is that on average, over time, as more edits and revisions are made to articles, the less biased they become. So, as English-language Wikipedia gets older, and the number of explicitly slanted articles trends downwards as the number of edits increases, should there still be concerns regarding bias?
What the above articles have shown is explicit, intentional bias. For all its difficulties to fix, it is usually visible and definable. What is less perceptible, especially in one’s own culture, is implicit bias – unconscious bias which shows itself in ways most people wouldn’t be aware of.
The vast majority of Wikipedia’s editors and contributors reside in the Global North, with Europe and North America having a ‘disproportionately loud voice’. This has led to multitudes more articles being geotagged in the North than the South, meaning topics regarding the South are much less visible over the internet. In 2021, for example, a study found that there were more articles geotagged in Paris on English-language Wikipedia than articles geotagged within the entirety of Africa.
This has also led to a geographical bias within the relatively few articles relating to the Global South. These topics often lack comprehensive coverage, and many are written from a limited perspective, sometimes containing inaccuracies or omissions – such as negating to include local cultural influences on the subject, opting instead for a breakdown from a Western “outsider’s” perspective. This creates a skewed representation of knowledge, where the experiences and topics of the Global South are marginalised and overlooked, which can perpetuate Eurocentrism across the internet. References on articles can be vastly unbalanced, with most coming from the Global North, especially from ‘the “ivory tower” of [Western] academic historiography’.
How could all these problems be fixed? Although there is no single solution, researchers David Laniado and Marc Ribé suggest increasing editor diversity, promoting awareness of implicit bias, and encouraging interactions between different Wiki languages and different regions and cultures. Only then, they argue, can Wikipedia fulfil its potential as a truly universal bank of knowledge.
George Orwell called these issues the ‘mutability of the past’ – where purposeful or not, individuals and institutions can rewrite history to fit certain versions of the truth. For all that it's good for and all the good it's done, the Wikipedia project has issues that must be solved. Inherent biases threaten its claim to neutrality and its ability to represent a truly global and inclusive knowledge base. While with time and inclusion some of these issues may be fixed, others seen unlikely to change anytime soon. So, maybe our teachers’ warnings were right all along – maybe Wikipedia isn’t as trustworthy as we thought it was.
Image courtesy of Kolossos via Wikimedia, ©2013. Some rights reserved.
The views and opinions expressed in this article are those of the author and do not necessarily reflect those of the wider St. Andrews Foreign Affairs Review team.
