James Madison could not have envisioned the internet, but he certainly anticipated the threat of misinformation and polarization to a society.
In “Federalist No. 10,″ Madison called it the power of “faction,” or extreme partisanship, that “inflamed [men] with mutual animosity” at the expense of the common good. Madison, however, thought that the sheer size of the United States and the deliberative, moderating aspects of the Constitution might make it more difficult for anyone to spread outrage widely beyond their states.
Enter technology and its amplifying impact on human instincts.
In the 1800s, the telegraph connected far-flung communities with information that would take weeks, if not months, for messengers on horseback to convey. Likewise, since the late 1990s, the internet has transformed every aspect of society, from the way people do business to how they find information or interact with each other. And as never before, the enormous influence of computer algorithms, artificial intelligence and other advances reflect and shape preferences, and transform conversation into conflict, facts into baseless opinions, and tolerance into intolerance.
Dissent and free speech are fundamental to a functioning government and society, but polarization is anathema to those principles. Most Americans believe in a society of principled consensus and respect for the processes designed to bring us together and shield us from the depths of extreme partisanship. In this editorial series, The American Middle, this newspaper is taking a serious look at the causes and solutions for a nation that has prioritized demonization over principled disagreements.
The internet and social media networks amplify communications beyond neighborhoods and oceans in ways that challenge the ability of human beings and political systems to discern fact from fiction and react appropriately. Millions of websites and pages are accessible from a smartphone, often with no indication of the veracity of the source.
Researchers who have studied these trends — and there are many — don’t agree on a single cause or the level of influence technology, social media and the internet play. Researchers generally agree, however, that hyperpartisan radio, cable TV outlets, social media, sharing platforms like Facebook, Twitter and YouTube, and the insidious power of friends sharing misinformation combine to advance a perfect storm of chaotic, angry and uncivil online exchanges.
So what happened since the early 2000s when compromise wasn’t a dirty word and tech startups like Friendster, Myspace and Facebook came into existence as novel ways to connect friends and present the most flattering depiction of our lives? Absent was a component of outrage that gradually emerged in the ensuing years as social media platforms developed. A mild communications medium among friends evolved into an engagement-driven automated online concierge that force fed information and enticed users to share the good, and especially the bad and ugly.
Researchers point to this as an indication that technology behind social media creates stereotyped portrayals of others that politicize otherwise nonpolitical issues, harden opinions and interactions, and unite like-minded users in echo chambers of misinformation and mistrust. And in many instances content-ranking algorithms not only provide like content but also suppress exposure to other news and contrary viewpoints.
To track such trends, academics are measuring affective polarization, which they define as the gap between individuals’ positive feelings toward their own political party and negative feelings toward the opposing party.
The methodology assigns values on a scale that ranges from cold or extremely negative (0 degrees), to neutral (50 degrees), to warm or strongly positive (100 degrees). In the 1970s, attitudes toward people holding opposing political views in the United States were around 48 degrees, or pretty neutral. In 2020, they measured an icy 20 degrees.
By 2016, Americans were rating their own party 45.9 points higher than the other party, an average increase of 4.8 points per decade in negative feelings toward members of the other party compared with one’s feelings about their own party.
In the social media world, success generates imitators and innovators with new social media apps and platforms appearing monthly. Reddit, TikTok and Instagram are the next generation apps, with no better reputation for enabling toxicity online.
More broadly, the result is stereotyped portrayals of others that politicize otherwise nonpolitical issues, harden opinions and interactions, and cocoon like-minded users in echo chambers of misinformation and mistrust. Jaime Settle, an associate professor of government at the College of William & Mary, writes in her 2018 book, Frenemies: How Social Media Polarizes America, that several Facebook features heighten divisiveness among users, and also drive users down rabbit holes of information and misinformation.
The 2016 U.S. presidential election campaign was the seminal moment for online disinformation. It was during that campaign that we learned of attempts by Russia and other foreign governments to influence the outcome of the presidential election in favor of Donald Trump by sowing confusion and misinformation. But that election also marked a sharp increase in Facebook and Twitter posts from elected officials, according to Pew Research Center. That trend continued and made the 2020 election more of an online contest than the preceding presidential cycle.
Pew also found two interesting trends. Lawmakers often did not include links to outside content. When they did, they engaged in what Pew calls “Link polarization,” meaning that they primarily or exclusively shared a handful of domains that affirmed their perspective. Pew’s research showed that 188 individual domains accounted for 62% of all the links posted by lawmakers during the last two election cycles, and that the number of these popular domains shared exclusively by members of one party — and not at all by lawmakers from the opposing party — increased from 20 in 2016 to 31 in 2020.
This isn’t just a political problem, it’s a social one, too. Medical experts say social media activates the brain’s reward center to release dopamine, a chemical linked to pleasurable activities such as sex, food and social interaction. The intersection of social media designs with human reactions makes for an addictive and potentially dangerous cocktail, especially to impressionable, developing young minds.
Unlike generations born before the 1990s, young adults today are growing up in a world dominated by smartphones and access to tweets, posts, likes, shares, TikTok challenges and influencers promoting idealized lifestyles online. Various studies tie heavy social media use to a rise in mental health issues such as depression, anxiety, diminished self-esteem, loneliness, memory loss, poor academic performance, nausea, headaches, muscle tension, tremors and suicides.
To further answer this question of how social media changes us, researchers followed a random sample of people whom they kept off of Facebook for a month before the 2018 midterm elections. Their results, published in the American Economic Review in 2020, found “significantly reduced polarization of views on policy issues,” but no diminishment of divisiveness based strictly on party identity.
In its own study on polarization, New York University’s Stern Center for Business and Human Rights pointed to work by Zeynep Tufekci, a sociologist and associate professor at the University of North Carolina, as an example of social media’s influence. Tufekci noted that her clicks on YouTube videos of Trump rallies prompted the platform’s recommend and autoplay features to recommend videos featuring white supremacist rants, Holocaust denials and other disturbing content.
Similarly, after she created a different account and searched videos of Hillary Clinton and Bernie Sanders, she received video recommendations of conspiracy theories. Among other things, they asserted that the United States government was behind the 9/11 attacks.
Even sharing information online has an impact. In work published in the Journal of Consumer Psychology, researchers at the University of Texas at Austin found sharing articles online, even without reading them, can cause us to assume an “expert” identity, making us overconfident of our knowledge and affecting our behavior. That is a significant observation. According to a Pew Research Center survey in 2021, roughly 72% of American adults, and 84% of adults between 18 and 29 years old use social media, many several times a day.
And the words we use in sharing matter too. In 2017, NYU researchers measured the reach of half a million tweets and found that each moral or emotional word used in a tweet increased its virality by 20%, on average. And Pew Research Center found in 2017 that posts exhibiting “indignant disagreement” received nearly twice as much engagement in the form of likes, comments and shares than other less critical posts.
Facebook, YouTube and other social media and sharing platforms have pushed back against such characterizations even as those who have been on the inside of social media companies are speaking out. Frances Haugen, a former Facebook data scientist and whistleblower, told a Senate subcommittee last year that Facebook’s profit motive harms children, encourages division and undermines democracy.
The pushback is bipartisan. Sens. Richard Blumenthal, D-Conn., Josh Hawley, R-Mo., and Elizabeth Warren, D-Mass., are among a growing list of ideologically diverse elected officials who have called for a crackdown on what they say are dangerous business practices.
So how should we think about social media in the decades going forward?
It is now far clearer than a decade ago that posting and forwarding links, images and videos can have a negative impact on other people and on a politically polarized society. But it is also apparent that technology has outpaced regulation and tugged at the fabric of a democracy that protects free speech.
That free speech protection has never been absolute, and Congress has acted historically to regulate communications and media in ways that conform with the First Amendment. The last time Congress overhauled communications law was the Telecommunications Act of 1996, which replaced the moribund Communications Act of 1934.
The sweeping measure was designed to encourage growth and entry into the newly deregulated communications businesses, an essential next step after the courts broke up the old Ma Bell monopoly that had blocked head-to-head competition for customers and innovation. But in 1996, there were no smart phones or social media, the internet hadn’t yet become a part of every person’s life, and advertising business models had not been built on digital user interactions.
No one can put the genie back into the bottle. However, NYU’s Stern School of Business, which has studied polarization for several years, concludes that reducing radical divisiveness is possible.
At the top of its list of proposed reforms is for the federal government to create and enforce new standards for industry conduct, such as requiring social media companies to disclose far more about how their algorithms rank, recommend and remove content. And social media platforms need to find alternatives to current business models that reward users for viral posts. Reducing the importance of “likes” and “shares” might encourage consideration of content on its merits, rather than on whether it provokes outrage, hatred or fear.
Another part of the legal landscape is Section 230 of the Communications Decency Act of 1996 which protects online platforms from civil liability based on third-party content as well as provides a measure of immunity for removal of content in certain circumstances.
Specifically, Section 230 says that “No provider or user of an interactive computer service shall be treated as the publisher or speaker of any information provided by another information content provider,” a powerful legal shield that unfairly protects tech giants from the legal consequences for hosting or transmitting libelous or threatening comments. Newspapers and broadcasters don’t have this protection. Reforming Section 230 protections, especially as it relates to social media companies, would begin to restore a measure of civility online and diminish the reach of algorithmically charged poison.
We would also ask lawmakers and the tech experts to explore ways to limit the influence of anonymous and unaccountable automated bots that appear to wield an outsized influence on online strife. Bots reinforce and exaggerate user counts and other engagement measurements such as likes, click-throughs and retweets; create fake engagements; influence the algorithms; and distort the intensity of the popular content that users receive.
Platforms like Facebook, YouTube, Twitter and newer social media iterations like Reddit are not the only causes of political polarization, but they do make it worse. For the sake of civil discourse, our national fabric and a healthy American Middle, the platforms that sow polarization must be part of the solution.
A better discourse
Madison’s warning that extreme polarization could erode our democracy seems to ring truer each day. While partisan differences are natural and even healthy, discourse that doesn’t allow for the common good to emerge is destructive. And at times it is an uncomfortable truth that we sow the seeds of our own destruction when we accept extremism as normal discourse and allow the tools of the modern world to amplify its reach and influence.
Modern technology has created important tools that we didn’t fully understand, and require a serious reevaluation and new regulations and protocols to encourage conversation, not hostility. Speech is powerful and must be respected for giving participants a meaningful voice, something authoritarian societies overtly ban.
However, free speech is not the same thing as an unregulated commercial communications system. Absent the bounds that can encourage the civil exchange of ideas, we are left with unproductive rants and malignant influences on politics, culture, national security and democracy. It is the reason we have speed limits and rules of the road and guardrails on all sorts of human interactions.
Our nation is strongest when it allows disputes to be peacefully worked out and trusts core institutions. Restoring The American Middle depends on social trust and civil discourse as the foundation of constructive civic engagement.
Part of our Opinion series on The American Middle, this editorial examines the way that technology has contributed to political polarization and what we might do to promote fairer, less adolescent public discourse.
We welcome your thoughts in a letter to the editor. See the guidelines and submit your letter here.