Election 2020: The impact of social media -
Close Menu
Close Menu

Election 2020: The impact of social media

Since at least as far back as ancient Greece, politics has been a public spectacle. In the United States for about the last decade or so, this spectacle has been updated to include social media. A seemingly endless river of minor posts and superficial reactions – often outrage – has flooded our digital devices and minds, often pulling us harshly in multiple directions.

Professor Benjamin Peters smiling, wearing eyeglasses, a collarless black shirt and a grey blazer
Professor Benjamin Peters

The University of Tulsa’s Hazel Rogers Associate Professor and Chair of Media Studies Benjamin Peters is an expert on the role of global media and national politics. He has written broadly on media, including a recent piece on how to use social media before and after an election.

Now, in the final days before one of the most momentous federal elections in recent U.S. history, Peters shares his thoughts via a Q&A on the complex ways Facebook and other social media platforms are being used to shape and contest the 2020 election, including the unsettling attempts of foreign adversaries to undermine the public’s confidence in the process itself.

How have media — and social media, in particular — shaped elections in the past and are likely to shape the current contest?

All elections are unprecedented, but this election is perhaps a bit more unprecedented than most. All elections have unusual conditions, but this one — caught up in a global pandemic — stands out for being especially unusual.

The pandemic has thrown most of our interactions online rather than face to face. This fact has given social media an additional degree of reach and power in 2020 compared to earlier elections. At the same time, however, it is vital to stress that most of our favorite hypotheses about social media — social media will throw the election in one direction, social media is destroying or saving American public discourse — will ultimately miss the mark.

History rhymes, not repeats. The past patterns we see in election media do not determine, but they can influence and help predict what is happening in this one.

We should try to imagine subtle ways that social media will affect the election rather than imagine the election, or democracy, will be run roughshod by unanticipated social media influences.

For example, we know from marketing and advertising research dating back to even the mid-1950s that certain people — or influences — often have an outsized influence on those in their social circles. Whether recommending the new Tide soap over the neighbor’s fence or penning the next viral tweet, these “influencers” are not media moguls, but they do possess influence among their friends and supporters. Today, micro-celebrities and influencers, thanks to the scaling of social media, can potentially have global reach; although in fact, of course, the vast, vast majority of social media influencers do not.

Are all social media platforms equal in terms of their political influence? 

Social media as a single label should often be avoided, because there are a great many differences between each and every one of the platforms. For example, Instagram cultivates a well-honed, edited aesthetic while TikTok rewards a more casual, relaxed vibe. But, like everything else, these are both carefully crafted. The echo chamber of a Facebook group is not the same political beast as a group of friends on Snapchat or a bunch of pseudonymous randos on Reddit subthreads. Each platform deserves its own analytic approach.

Still, it is true that, generally, social media does have a democratizing effect: all people online will now have a bit more voice than before. But it would be a fundamental mistake to imagine that social media will extend equal voice, or equally worthwhile voices, to all people. Social media, we should remember, are plural as are their effects on politics.

How do you think the mainstreaming of social media has changed the tenor of political discourse in the current presidential election? 

There are articulate, sophisticated thinkers on all sides of elections; that said, many Trump supporters on social media, especially on Facebook, favor political discourse that highlights mostly snappy slogans, meme-length attention spans and the call and response of large rallies and crowd chants. Social media does not demand meme-length political discourse, but it certainly does reward it, and the result, unfortunately, has meant a striking impoverishment of political discourse.

This impoverishment proceeds in at least two different directions at the same time: meme-length reductions and conspiratorial fringes.

Thanks to social media, it is easier for those with extreme views to find small groups of people who will listen to those views — say, about Satanic pedophile rings, baby-part trafficking or 5G networks. Many of these groups eventually form into loose-knit conspiratorial movements that, perversely, take confidence in the fact that outsiders do not share such special “knowledge.” QAnon is only the most recent example of such knights of non-knowledge.

For this reason, dismissive claims of “fake news,” “political correctness” and even accusing others (no matter how justifiably) of harboring “conspiracy theories” may reinforce the motivated reasoning of those with extreme views on social media. Perhaps the greatest social media culprit is the Facebook Group, a self-moderating echo chamber. People enter only out of self-interest and remain only if a like-minded moderator approves their behavior. It is a worrying canary in the mineshaft if the only source of opposition one encounters online comes from a troll.

How likely are foreign disinformation campaigns conducted via social media to influence the current election? And how would we know if they are successful? 

It is tempting to blame the results of an election on some foreign power, but it is important to realize that in the case of the 2016 election, the influence of Russian disinformation campaigns was apparently very small. Thankfully, in 2020, there have been more steps taken to prepare for foreign influence, so we’re on the lookout for tactics used in 2016’s tampering.

For example, in 2016, Russian tampering took advantage of the partisan divide that is present in American politics, simply by indirectly contributing to and magnifying the already native internal dissension, division and fighting not among the political parties but about the political system itself. It is vital to draw a distinction here: political debate between the parties is supremely worthwhile; debate is the lifeblood of what makes our democracy work. However, those who spread doubt about the integrity and legitimacy of the system itself — our electoral process, the value of voting, the need for calm, informed disagreement — are serving our country’s opponents.

Don’t blame the dark heart of America on foreign powers. Unfortunately, election tampering is as old as the idea of elections, and America, in particular, has an embarrassing history of meddling in elections in foreign countries. No country has done more of it than America, in fact.

So, the fact that we are facing the threat of election tampering from both foreign and potentially internal sources in 2020 is not surprising. So far, it appears that most foreign powers’ attempts to meddle are largely unsophisticated: Russian hackers, thus far, have proven to be less surgeons of social media and, rather, blunt-force bards who spin tall tales and sow doubt for dimes that are fairly obvious to the educated and media-literate classes.

Avoid the temptation to scapegoat a foreign power. It is too easy. It excuses us from the hard work of political reform and self-reflection that America needs. Instead, hold our own politicians accountable. Check and then get out the vote for the sources of verifiable corruption, lying and political vice here at home.

In the U.S, voting is a constitutional right. Find your local polling place, stay informed and cast your vote on or before Election Day on Nov. 3.