The febrile pitch of digital voices on all sides of Elon Musk’s purchase of Twitter the past few weeks seemed timed to confirm Jonathan Haidt’s recent thesis that social media is designed to deeply fracture us. Haidt’s widely-read essay in The Atlantic describes how social media has “magnified and weaponized the frivolous,” wrecking trust in institutions along the way. Haidt’s critique, similar to other recent studies, helps us understand how social media is designed to amplify passions and ideological extremes in an era in which “outrage is the key to virality.” 

Part of the reason passionate tribalism and ideological hyperbole drive fragmentation is because of another factor that gets less attention: the absence of moral sentiments in social media. Moral sentiments – the affections and inclinations that prompt us to do good – keep our passions from overwhelming us and our minds from misleading us. They are the product of inculcation over time. They steer us in the right direction without being taskmasters. They develop through repetition as we emulate and admire good behavior around us. Our hearts warm when we see someone pick up the groceries an elderly shopper dropped in the store or help a lost child find her parents. When we emulate that behavior, the rewarding nature of the experience prompts us to do it again, and so on. When we speak of our conscience or a “sense of duty” or say “I was compelled” to do this or that, we are usually talking about our moral sentiments.  

Think of the moral sentiments as that part of you between the passions in your gut and the reason in your head. Our passions sometimes lead us to do heroic things, like acting courageously in the face of sudden danger, but left to themselves they usually just get us in trouble. “Go with your gut!” is bad advice for someone whose moral sense is poorly developed. Likewise, our rational powers can explain the right course of action in this or that situation, but our sentiments – which we sometimes call our “heart” – are the gentle force that prompts us to act accordingly. 

C.S. Lewis described the same phenomenon when he wrote in The Abolition of Man: 

The head rules the belly through the chest—the seat, as Alanus tells us, of Magnanimity, of emotions organized by trained habit into stable sentiments. The Chest—Magnanimity—Sentiment—these are the indispensable liaison officers between cerebral man and visceral man. It may even be said that it is by this middle element that man is man: for by his intellect he is mere spirit and by his appetite mere animal.

Social media creates interpersonal engagement with little to no room for moral sentiments to act or grow – or, in Lewis’s formulation, little room for us to be truly human. It allows us to express passions immediately or articulate (often flawed and spurious) arguments, but it is a poor cultivator of moral sentiments, which require sympathy, reciprocity, and repetition. 

Philosophers since at least ancient Greece have been trying to resolve the problems of unbalanced passion and distorted rationality, and now we have created digital tools designed to amplify the worst of both. We use them to shout at our opponents, argue without the arts of persuasion, make our friends envious of our foodie experiences, and display charts and graphs proving how totally right we are. Designed for performance and expression with the expectation of immediate feedback, social media has unwittingly created a new way for us to engage with little to no room for the tempering effects of moral sentiment. 

Moral sentiment theory was first developed as a school of thought (with some important antecedents) in 18^th century Scotland by a cadre of writers including but not limited to giants such as David Hume and Adam Smith, as a pushback against overly rationalistic and metaphysical explanations of virtue. These writers were primarily interested in what we could observe about moral judgments and behavior, and it became evident to them that affection for doing the right thing develops by repeating and emulating good actions that are praised and embodied by people we respect. A parent (or at least a good parent) tells her children to make their beds not for the sake of the activity itself but to cultivate in them an inclination toward cleanliness and order. When people we admire praise noble actions, our self-interest compels us to be praiseworthy, too, but hopefully those praiseworthy actions eventually become more like a habit – something that “comes from the heart.” 

Social science research shows that the Scottish writers were onto something and that our good behavior is determined more by moral feelings than raw instinct or ideological belief. We act for the good (or bad) of others because of moral emotions that prompt us, not from self-conscious obedience to clear, calculating reason. James Q. Wilson, in his book, The Moral Sense, documented the numerous 20th-century studies that validated moral sentiment theory’s account of how we develop sympathy for others and feel a sense of duty to do what is right. Wilson observed “how well the account given by some modern scientists agrees with [Adam] Smith’s ‘prescientific’ view,” and accounted for the important role that families and culture more generally play cultivating and influencing how moral feelings develop. 

Because it cannot recreate the environment in which humans connect more deeply, social media is a poor vehicle for the cultivation and transmission of moral sentiments. As Haidt and others have shown, social media is more likely to cultivate the opposite of moral sentiments by creating intense interpersonal engagement without moral connectedness. The addition of tools such as retweet and “like” buttons, Haidt argues, has placed a premium on the outrageous behavior that best drives virality, and conversely, are ill-suited to the kind of interaction required to improve our moral habits. It is as if social media were purposely designed, using findings from the best psychological research, to create a space where the passions and ideological confusion overwhelm moral formation – where head and gut overwhelm the heart.  

But not all of life happens on social media, which might point towards a path forward which social media pessimists sometimes miss. The institutions that shape moral sentiments still exist, and it may be possible to reintroduce moral sensibilities into institutions that have been overwhelmed by social media. The informal and local institutions of daily life are the primary arena of moral formation, and a natural bulwark against passionate and ideological fragmentation.

For instance, data from an oft-cited survey by the organization More in Common (which Haidt quotes in his essay) show that a majority of Americans are not inclined toward extreme views (and presumably the behaviors that support them). My survey research with colleagues at the American Enterprise Institute has similarly found that most people prefer normalcy over extremes. There is also evidence that people with real friends in their neighborhood engage in a wide range of healthier social interaction – such as volunteering, having people over for dinner, even talking to strangers – than those who only have abstract notions of the goodness of others. We also found that the only volunteers in America who are lonelier than the national average are those who restrict their volunteer activities to politics, which is also a form of volunteering that people can engage in online or without working together with others. Most other volunteers are actually in their community – whether in their church, local charity, or sports and veterans clubs – and they are measurably happier, more optimistic, and less prone to ideological manipulation.  

Life that happens off social media is the best buffer against its negative effects, and it is probably the best explanation for why recent research shows that it may not be harming us in our personal lives as much as its critics suggest. If social media is an occasional platform for people who are also engaged in their local community, where the cultivation of moral sentiments happens, its effects will be limited. 

But what about our formal institutions, which is what Haidt and most social media critics worry about? As wounded as they are, our formal institutions can still serve as cultivators of moral sentiment despite the incentives in the opposite direction. Conservatives often understate the value of politics as a moral force, and progressives often overstate it, but our Constitution has always presupposed that morally sensible people would not only lead our political institutions but be shaped by them. Speeches, essays, appearances on television and radio, and meetings around conference tables can still affect the sentiments in ways that social media’s default impetus toward rapid feedback and virality have made nearly impossible. They may not be as effective as a parent, teacher or coach, but their impact is still very real. We must use the tools available to us to build up a healthier culture. 

Political leaders have an opportunity to make more space for moral suasion as a way to dial back the emotional and ideological extremism in our politics. Our political parties have the ability to emphasize the importance of persuasion again and not to elevate the candidates that excel in outrage. Committee hearings in Congress can be about governing again and not merely stages on which members act out the outrage and hyperbole for use in social media and campaign commercials.  

The difference between these reforms and those social media critics such as Haidt suggest – which include giving regulators the power to limit the virality of social media posts – is that there are no predetermined private sector lobbyists. The only barriers to political parties and Congress reforming themselves are themselves – and the people who support and vote for them.