It’s often said that social media has a polarizing effect on our politics. And, on the surface, this narrative makes a lot of sense. The polarization of politics has continued as social media has taken over our brains. And what social media does, among other things, is make a game of earning the approval of your peers, thus solidifying your group identity. When you post something that pleases the sensibilities of your cohort—whether it’s a handsome selfie or a solemn plea for stricter gun control—you get the satisfaction of an immediate bombardment of friendly notifications. The reward structure of the social media experience doesn’t provide incentives for expressing minority views, or objecting to the prevailing narratives, or befriending those who disagree with you.
Moreover, Twitter and Facebook aren’t great places for dialogue. Political arguments are usually futile in real life, even with all of the felicitousness provided by face-to-face interaction. It’s much worse when ideological disagreements need to be reduced to 280 characters, or haveto compete with cute pictures of somebody’s baby. In this setting, sensitivity and nuance doesn’t play well. What gets the most attention is pithiness and aggression. In short, social media enables the self-congratulation and self-separation of mutually hostile political factions. Sounds pretty polarizing, right?
Yes. However, there’s a big and obvious question here, which is whether this is actually any different from the pre-Twitter media landscape. Long before Facebook was ever a gleam in Mark Zuckerberg’s eye, the various political classes selected the media that was most collegial to their respective worldviews. To take America as an example, in previous decades, Christian conservatives tuned into right-wing talk radio to hear about the horrors of the gay agenda, whereas elite liberals picked up Harper’s to read about the horrors of capitalism. (This is still true today, in part.) Bubbles and echo chambers exist in absence of Twitter. All that’s required to create ideological homogeneity is tribal self-selection or homophily—the tendency of people to hang out with people who are like them and agree with them, given freedom of association. That’s definitely a pre-iPhone tendency.
But, of course, it’s still possible that social media has enhanced tribal patterns of behaviour—that this is not a difference of kind, but it is a difference in degree. So, if we check the data, what do we find? Well, it appears that social media does, in fact, have an effect on polarization. It’s just the opposite effect that critics might expect. According to a demographic study by Boxell et al., published by Stanford University, political polarization is actually less pronounced among demographics that use social media more often (young people, essentially). This shows that it’s unlikely that social media is a more powerful driver of polarization than old-fashioned media. (Or it shows that, even if social media does polarize, there’s some countervailing anti-polarizing force that’s much more powerful.)
And, like the just-so story about why social media polarizes, there’s an appealing readymade narrative about why the opposite might be true. While political disagreements on Twitter and Facebook tend to be shallow and nasty, they’re still genuine disagreements—something that doesn’t usually occur in traditional media. The New York Times doesn’t contain a second page declaring that all the articles on the front page are slanted. And while it’s true that debate programs are a staple of political television, such programs are usually staffed by a preexisting team who are paid to perform a predictable set of reactions to ongoing affairs. Meanwhile, on Twitter, it’s quite easy to run into novel objections to everything you believe in, which, even if they aren’t particularly convincing, might compel more considered private reflection.
Or maybe it’s even simpler than that. It’s possible that young people are less polarized because social media is so nasty and tribal. While a minority of social media influencers make a lot of provocative noise, it’s possible that the non-contributing majority is quietly alienated by the vitriol. While a controversial tweet with 1200 retweets looks impressive, there’s no way to measure the number of users who have quietly rolled their eyes and moved on—or have simply quit Twitter altogether.
There’s a larger lesson here, which is that it’s unwise to infer narratives of societal change based simply on the most visible behaviour provoked by one app or another. (Another demonstration of this: millennials have way less sex than their parents, despite the existence of Tinder and all the moral panic surrounding it.) Ultimately, sensationalist narratives about the polarizing effects of social media are just the kind of thing that’s popular on social media.