The hidden bias of internet memes

A meme posted by Amber Rose referencing her feud with the Kardashians.

A meme posted by Amber Rose referencing her feud with the Kardashians.

Published Apr 8, 2015

Share

Washington – Know Your Meme, the preeminent authority on all things viral, has documented some 12,302 memes since launching in late 2008.

There are silly GIFs. Ever-changing image macros. Too many in-jokes, YouTube clips and squash-faced cats to analyze or name.

And yet, in a new study published in the Journal of Computer-Mediated Communication, a team of Israeli researchers has managed it: They’ve conducted an actual census of Internet memes.

Now, before you get too excited, we should caution that the researchers, led by Tel Aviv University’s Elad Segev, didn’t actually tally every meme on the Internet. (That would be impossible. Also: crazy.)

But the researchers, who are interested in how ideas morph and branch as they move across the Internet, did analyse what they call the 50 most popular English-language meme “families,” which include the original meme (think: the very first illustration of David Silverman, captioned “are you serious?”) and its most widely circulated derivatives (all the “Seriously Guys” that came after).

“Analyzing,” in this case, involves trying to pin down all the things that make this meme a meme: what themes it covers, what physical form it takes, whom it does or does not depict. (Fun dinner party fact: “quiddity” is the collective word for all this.)

Once they’d analyzed some 1000+ memes that way, the researchers were able to calculate their most common features and map them according to similarities. And while we tend to think of memes as fluffy, universal – even meaningless! – things, they actually found some complex structures and biases baked into them.

To wit: Of memes that show people, versus dinosaurs or cartoons or cats, men appear twice as often. And nearly 45 percent of all the people in memes are Caucasian; Hispanic subjects make up a fifth of one percent, by comparison.

“These findings corroborate many of the observations made in (past) qualitative studies,” the researchers sum up, “in which the memetic sphere was described as dominated by young, white men.”

On one hand, of course, this conclusion isn’t particularly jarring: Both Reddit and 4chan, the traditional petri dishes of meme culture, are overwhelmingly young and male; Tumblr, a more recent meme-making upstart, is split about 50-50 on gender, and skews young across the board.

But even if these conclusions seem intuitive, they express a great deal not only about inclusion and diversity in online spaces, but about power and information in them: who the gatekeepers are, who determines what’s cool, whose instincts and interests are considered most “fit,” in the Darwinian sense of the term.

(Memetics, the study of memes, comes straight from evolutionary biology: Successful memes, like successful organisms, need to reproduce on blogs and forums and cheap T-shirts in order to survive.) The Internet, despite its more utopian goals, replicates biases and social structures that exist offline.

Of course, the Internet is a complicated place, and further research will be needed to get into all that; it might be worth exploring, for instance, memes that subvert mainstream biases or promote minority identities and issues, like “Successful Black Man” or 2012’s “Binders Full of Women.”

This study found that the further meme families fall from the mainstream, the more they stay on-theme – an intriguing hint at the dynamics in niche Internet communities.

Still, one thing has become pretty clear already: Memes, counter their rep, should be taken seriously.

Washington Post-Bloomberg

Related Topics: