It’s Not Filter Bubbles That Are Driving Us Apart

When people come face-to-face with what others believe, they may not like it.

An illustration of bubbles with shouting faces drawn on them
Getty; The Atlantic

In 1948, Sayyid Qutb went to America, where he stayed for almost two years. It was a formative experience for the devout, uptight, disgust-prone Egyptian, who is regarded as one of the founding fathers of Islamist ideology. To say that Qutb didn’t take to the place is an understatement: He thought it was soulless, materialistic, crass, haughty, and sexually permissive.

It is an exaggeration to suggest that America radicalized Qutb, but it does seem to have strongly reinforced his aversion to the country and the West in general, which he came to regard as an existential threat to Islam. What it categorically didn’t do was make him more moderate in his thought and sentiment.

Qutb’s American sojourn instantly came to mind as I read a new academic paper that seeks to understand the intensely polarized state of our social and political life. The paper, by Petter Törnberg of the University of Amsterdam, was published in October in PNAS, the official journal of the U.S. National Academy of Sciences. Its core claim is that “it is not isolation from opposing views that drives polarization but precisely the fact that digital media bring us to interact outside our local bubble.” Although Törnberg doesn’t mention Qutb, the process whereby antipathy and distrust toward others is intensified not by distance from those others but by direct contact with them, albeit filtered through existing cognitive biases, is strikingly borne out by Qutb’s American experience.

As Törnberg explains at the paper’s outset, the dominant account for understanding our current polarized condition holds that the internet, social media, and online algorithms have combined to divide us into warring tribes whose beliefs, identities, and mutual enmities become ever more entrenched in echo chambers, where groupthink dominates and competing viewpoints are banished. Many social scientists argue that this not only undermines democracy, which depends on the open exchange of ideas, but also serves to foment conflict, even outright violence. As the legal scholar Cass Sunstein put it, “Particular forms of homogeneity can be breeding grounds for unjustified extremism, even fanaticism.”

Törnberg’s paper rejects this account, arguing that far from sheltering people from opposing ideas and ways of thinking, digital media has, in fact, served to “bring us to interact with individuals outside our local bubble,” where many interactions take on a warlike character and “we are forced to take sides.” Our main problem, as Törnberg conceives it, is not that we spend too long listening to the comforting voices on our own side, but rather that we’re too attentive to the loudest, most enraged, and most unhinged voices on the other side.

To better grasp the paper and its wider implications, I recently spoke with Törnberg. As a fellow researcher who has studied audience engagement with online atrocity propaganda, including jihadi beheadings and other unmentionable cruelties, I was particularly keen to ask him about the problem of distortion and how overexposure to extreme online material can warp people’s view of the world so that, in a reverse process of desensitization, they become ever more alert to premonitions of catastrophe and societal collapse.

“On mainstream social media, we don’t find so many echo chambers,” Törnberg told me, adding that “there’s plenty of interaction going on.” More crucially, he said, “that interaction doesn’t consist of rational arguments that lead to moderation, that’s just not how it’s playing out.” In his view, many of our online interactions are driven not by good-faith initiatives to better understand one another, but by a tribal imperative to signal our moral superiority over our partisan enemies—especially if members of our own in-group are watching. This is usually done through mockery or vilification with little or no regard to the rules of civilized discourse.

These dynamics are readily apparent on Twitter, and on many other social-media platforms, where what dominates is not protective disengagement from opposing viewpoints but an ecstatic clamoring for their most extreme articulations. Political partisans reinforce their sense of moral identity by seizing on these articulations and excoriating them, vividly marking out the parameters of good and evil, while firmly placing themselves on the side of the gods. Partisans are motivated to do this, in part, because it’s so emotionally rewarding: One shouldn’t underestimate the pleasure of feeling indignant and righteous. But it’s also, as research has shown, the key to going viral: Inflamed or “moralized” posts about political opponents are substantially more likely to be shared on social media. And one really shouldn’t underestimate the pleasures of mass endorsement on social media, either.

Before the rise of the internet and social media, most people were relatively secluded from their most extreme political adversaries; a person might read about those adversaries in a newspaper or see them on TV, but didn’t have access to their personal lives. Now anyone can watch them daily on TikTok, voyeuristically marveling at their outrageousness. Because there is no shortage of mentally unwell people who are willing to expose themselves on social media, there is always a self-replenishing reservoir of political derangement for activists in all places along the political spectrum to seize on for their ends.

The social psychologist Jonathan Haidt has for years observed this dynamic. “So long as we are all immersed in a constant stream of unbelievable outrages perpetrated by the other side,” he told Vox in 2018, “I don’t see how we can ever trust each other and work together again.” Törnberg voiced a similar concern to me, referring to how social media has encouraged an “outrage-driven politics” that crowds out productive democratic engagement.

When Sayyid Qutb went to America, he was exposed to a range of people who thought very differently from him. Instead of prompting him to critically reflect on his own beliefs and convictions, this seems to have led him to do the opposite: He seized on what he saw as the most outrageous forms of moral perversity he encountered and considered them indicative of America and the West as a whole. When he returned to Egypt, he then sought to disseminate this one-sided account among his followers. The historian John Calvert wrote that Qutb viewed the United States “not with fresh untainted eyes, but rather through the tinted spectacles of a man long captive to a particular view of the world.” He “would either purposefully ignore or simply not see anomalies which contradicted” his account of what America was.

If Qutb’s intellectual failings sound familiar, that’s because we are all too prone to them ourselves. What the rise of social media has done is make them markedly worse, as political derangement is now everywhere, warping our sense of proportion and judgment. It’s hard to know what to do about this, in large part because the appetite for it is so strong. As former President Barack Obama said in April, “There is a demand for crazy on the internet that we need to grapple with.”

The really striking thing about Qutb’s American experience is how transfixed he was by what he regarded as American permissiveness. Not only did he have an uncanny knack for finding it, but he couldn’t keep off the subject when he did run into it. Perhaps he enjoyed the feeling of moral disgust and the sense of righteousness it gave him, or perhaps he was secretly attracted to what America had to offer. Regardless, if we’re to properly inquire into how we’ve become so polarized, we had better start thinking about that demand for crazy—and what the people who consume and share that crazy are getting from it.