Ariel Procaccia is Gordon McKay Professor of Computer Science at Harvard University.

In the run-up to the 2020 election, mail-in voting was taking a public battering, with allegations of impending fraud from one side of the political aisle and cries of intentional delays by the U.S. Postal Service from the other. Yet at the same time, a much more benign process was playing out in my social media feeds: They were inundated by people’s proud announcements that their mail-in ballots had been received and recorded. Americans were voting with their tweet.

Beyond their role as a comforting escape from the horrors of traditional election coverage, there’s a case to be made that such posts may have contributed to the election’s record turnout. It’s one way in which social media — rightly criticized for its part in increasing polarization and spreading disinformation — may have helped democracy.

At first glance, this claim probably seems counterintuitive. The 2020 turnout was undoubtedly bolstered by strong feelings on both sides. One would think that the pandemic-driven expansion of mail-in voting also contributed to turnout by making it easier to vote — a plausible explanation that has nothing to do with social media. The problem with this reasoning is that there’s plenty of evidence — for example, from the United States and Switzerland — that mail-in voting has historically had a very limited impact on turnout.

This phenomenon is elucidated in a remarkable 2010 paper by political economist Patricia Funk. Mail-in voting’s obviously positive effect of reducing the effort required to vote, she argues, is counteracted by the negative effect of simultaneously reducing the social pressure to vote. Funk posits that one of the reasons people vote is to be seen by neighbors as they perform their civic duty; by contrast, mail-in voting “renders the voting act unobservable,” thereby taking away this social incentive.

But how can such a hypothesis be tested? Funk points out that social pressure is known to be strongest in small communities, where people are more likely to gossip about their neighbors’ virtues and vices. If the desire to be seen voting did significantly contribute to turnout, we would expect the introduction of mail-in voting to have a less positive impact on turnout the smaller the community is. Funk finds that this is indeed the case by analyzing several decades’ worth of federal elections in Switzerland, where different cantons (the Swiss equivalent of states) introduced mail-in voting at different times. In fact, while overall turnout slightly increased with mail-in voting, in certain small communities the option to vote by mail decreased turnout.

Nowadays the online world is home to millions of small communities; Facebook, Instagram, Twitter and other platforms have become social pressure-cookers. Especially in this pandemic-burdened time of homemade bread, homemade haircuts and homemade ballots (of the legitimate variety), social media gave America a socially distanced opportunity to make voting observable again. And voters could influence their friends to vote too, while they were at it.

In 2020, a time-honored system of democracy may have thrived, against all odds, thanks in part to social media. It would be just another irony in a year that has had more than its fair share.

Read more: