Who today doesn’t know the power of hashtags in social media sites like Twitter and Instagram, and how they can isolate your content and streamline your information to circle around a single word. Instantaneously you are able to filter in only tweets and posts related to that word.
What seems like an ingenious strike for most social media users who might identify themselves, their lifestyle, or perhaps their mood with a single word or hashtag, became a matter of controversy for the providers of Instagram.
The controversy of moderating controversy
In 2012, Instagram started regulating certain words used by people who were in favor of anorexia — a type of eating disorder, which over long periods can lead to severe health issues. With its move to tie down certain words, Instagram became one of the “more tightly moderated social media platforms.”
Researchers from Georgia Tech decided to study the matter, and determine whether the restriction and ultimately banning of words actually made any difference in the communities using them.
A research paper issued by the group states that banning certain words just made the users of those words linguistically more creative in working around the banned spelling or substituting the topic describing words completely.
And while the witch hunt initially began with a ban of 17 terms, the pro-eating-disorder groups then started using several hundred terms specifically dedicated to these pro-anorexia groups. The ban apparently even increased the desire to become part of this controversial group on Instagram.
“Our research has explored linguistic, behavioral, and topical changes in pro-ED [pro-eating disorders] communities in the aftermath of Instagram’s moderation of pro-ED tags in 2012,” according to the summary of findings in the research paper.
Instagram started vigorously regulating ” terms like, ‘thinghgap,’ ‘imugly,’ and ‘thinspiration’.” Searching for any term on the red-list of tags would either show no results if it was banned, or the user would have to view a screen informing them that they are about to view controversial content, and sometimes also providing links to helpful resources or hotlines.
Unfortunately all the moderating of keywords only seemed to lead to an increase in the number of variations members used to access the controversial community on Instagram.
“Moderation in April 2012 led to the emergence of lexical variants of banned tags and tags with advisories. In fact, lexical variation showed a monotonic increase over time, indicating a desire on the part of the community […]
“Thus, content moderation has been mostly ineffective at decelerating the dissemination and proliferation of pro-ED behavior on the platform. ” the summary of findings continues.
One pitfall of moderating a community’s search terms is that this action might result in causing a type of camouflage effect through the linguistic adaptation of abstract terms to work around the censorship.
[…] If the community continues to move to increasing lexically variant tags, and away from the more common pro-ED tags, as our results show, it would be difficult to discover them and thereby report, remove, or bring help to them.”
With the advance of technology and the further integration of applications into personal life and society’s spiritual culture, the line between privacy, freedom of speech, and morality lead enforcement becomes quite foggy.